estimation of source area, river paleo-discharge, …...invited review estimation of source area,...

Earth-Science Reviews 153 (2016) 77–110

Contents lists available at ScienceDirect

Earth-Science Reviews

j ourna l homepage: www.e lsev ie r .com/ locate /earsc i rev

Invited review

Estimation of source area, river paleo-discharge, paleoslope, andsediment budgets of linked deep-time depositional systems andimplications for hydrocarbon potential

Janok P. Bhattacharya a,⁎, Peter Copeland b, Timothy F. Lawton c, John Holbrook d

a School of Geography and Earth Sciences, McMaster University, 1280 Main St., West, Hamilton, Ontario L8S 4L8, Canadab Department of Earth and Atmospheric Sciences, University of Houston, Houston, TX 77204, United Statesc Centro de Geociencias, Universidad Nacional Autónoma de México (UNAM), Blvd. Juriquilla No. 3001, Querétaro 76230, Mexicod School of Geology, Energy, and the Environment TCU, Box 29330, Fort Worth, Texas 76129, United States

⁎ Corresponding author.E-mail address: [email protected] (J.P. Bhattachary

http://dx.doi.org/10.1016/j.earscirev.2015.10.0130012-8252/© 2015 Elsevier B.V. All rights reserved.

a b s t r a c t
a r t i c l e i n f o
Article history:Received 25 September 2014Received in revised form 28 September 2015Accepted 27 October 2015Available online 31 October 2015

Keywords:Source to sinkSequence stratigraphyRiversDeltasSubmarine fansPaleodischargeProvenanceDetrital zirconsPaleoslope

The source-to-sink (S2S) concept is focused on quantification of the various components of siliciclastic sedimen-tary systems from initial source sediment production areas, through the dispersal system, to deposition within anumber of potential ultimate sedimentary sinks, and has more recently been applied to deep-time stratigraphicsystems. Sequence-stratigraphic analysis is a key first step that allows depositional systems to be correlated andmapped, within a time-stratigraphic framework, such that fluvial transport systems can be linked to down-dipshorelines, shelf and deep-water deposits and interpreted in the context of allogenic processes. More recently,attempts have been made to quantify catchment areas for ancient depositional systems, using scaling relation-ships of modern systems. This also helps predict the size of linked depositional systems, such as rivers, deltasand submarine fans, along the S2S tract.The maximum size of any given depositional system, such as a river, delta, or submarine fan, is significantlycontrolled by the area, relief, and climate regime of the source area, which in turn may link to the plate tectonicand paleogeographic setting. Classic provenance studies, and more recent use of detrital geochronology,including zircons, provide critical information about source areas, and place limits on catchment area.Provenance studies, especially when linked to thermochronometry also provide key information about rates ofexhumation of source areas and the link to the tectonic setting.In this paper the techniques for estimation of water and sediment paleodischarge and paleo-drainage area areoutlined, and sediment budgets are calculated for a number of Mesozoic systems, primarily from westernNorth America. The relevance for hydrocarbon exploration and production is discussed for each example.In Mesozoic Western Interior basins of North America, extensive outcrop and subsurface data allow the largesttrunk rivers to be identified, typically within incised valleys. Thickness, grain size, and sedimentary structurescan be used to infer slope and flow velocities, and using width estimations, water and sediment paleodischargecan be calculated. River paleoslope can also be independently measured from stratigraphic-geometric consider-ations and used to assess paleo-river flow. Paleodischarge in turn is used to estimate the size of the catchmentsource area. Paleodischarge of rivers can also be estimated independently by integrating estimates of catchmentsource area, for example by using detrital zircons integrated with paleoclimate.The catchment areas of North America evolved significantly during the late Mesozoic. During the Jurassic–EarlyCretaceous, fluvial systems consisted of continental-scale low slope (S = 10−4), axially drained rivers, formingthe 40-m-deep channels in the Mannville Group in Canada, which now host the supergiant heavy-oil-sands re-serves. During the times of maximum transgression of the Cretaceous Seaway, such as the Turonian and Campa-nian, the western North American foreland basin was characterized by smaller-scale (typically 10-m deep),steeper gradient (S=10−3) sand and gravel bedload rivers, dominated by transverse drainages in the rising Cor-dillera. This created a number of smaller river-delta S2S systems along the coast. As the Laramide Orogenyprogressed, the Western Interior Seaway receded, and by the Paleocene the modern continental-scale drainageof North America was largely established with a major continental division separating south-flowingMississippidrainages from north-flowing systems.The integration of paleodischarge estimates with provenance analysis enables the improved use of the sedimen-tary record tomake estimates about the entire S2S system, as opposed to primarily the depositional component.

a).

http://crossmark.crossref.org/dialog/?doi=10.1016/j.earscirev.2015.10.013&domain=pdf

http://dx.doi.org/10.1016/j.earscirev.2015.10.013

mailto:[email protected]

http://dx.doi.org/10.1016/j.earscirev.2015.10.013

http://www.sciencedirect.com/science/journal/00128252

www.elsevier.com/locate/earscirev

78 J.P. Bhattacharya et al. / Earth-Science Reviews 153 (2016) 77–110

Clinothem stacking relationships and isopach mapping of stratigraphic volumes have also been integrated withchronostratigraphic data to analyze long-term S2S sediment budgets. A more quantitative approach to estimat-ing the scale of erosional, transport and depositional components of sedimentary systems, especially in the con-text of linked source and depositional areas, also puts constraints on the size and scale of potential hydrocarbonreservoirs and thus has significant economic value.

© 2015 Elsevier B.V. All rights reserved.

Contents

1. Introduction: S2S in deep-time systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 782. Sequence-stratigraphy overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 793. Critical considerations in applying S2S concepts to deep-time systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 804. Components of S2S dispersal and depositional systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82

4.1. Alluvial systems, catchments and transfer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 824.2. Shorelines, shelves, and along- and across-shelf transfer of sediment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 844.3. Deep-water systems, the ultimate sink . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84

5. Source area versus discharge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 846. Paleodischarge, paleoslope, sediment budgets, and source-area estimates in deep-time systems . . . . . . . . . . . . . . . . . . . . . . . . . 86

6.1. Characterizing the trunk river . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 866.2. Estimating channel dimensions and sediment characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 866.3. Paleoslope estimates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 886.4. Velocity and paleodischarge estimates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 896.5. Paleodischarge estimates using regional hydraulic geometry curves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 906.6. Estimates of sediment budgets from paleodischarge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 916.7. Drainage area and reservoir quality from provenance analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 926.8. Use of detrital zircons for formation age determination and thermochronology (uplift and exhumation rates) . . . . . . . . . . . . 936.9. Tectonics, subsidence and sediment supply . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94

7. Examples of Mesozoic S2S systems in Canada and the USA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 947.1. Triassic–Jurassic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 957.2. Big rivers in the early Cretaceous . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 957.3. Late-Early Cretaceous, end of the big rivers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 967.4. Late Cretaceous (Cenomanian), rivers and deltas of the Dunvegan Formation and equivalents . . . . . . . . . . . . . . . . . . . . . . . 97

7.4.1. The Dunvegan delta . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 977.4.2. Dunvegan paleodischarge estimates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 977.4.3. Dunvegan sediment budgets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1007.4.4. Dunvegan petroleum perspective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100

7.5. Turonian: a time of many deltas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1017.6. The Ferron sandstone, a case study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1017.7. Santonian to Paleocene, the mega-bajada . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103

8. Conclusion and future work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105

1. Introduction: S2S in deep-time systems

The source-to-sink (S2S) concept focuses on quantification of thevarious components of sedimentary systems from source areas, throughdispersal systems, and thence to deposition in a number of potentialsedimentary sinks (Fig. 1). Boundaries of particular interest includethe gravel–sand transition in the alluvial realm, the shoreline, theshelf-slope break, and the continental slope to deep-water transition.The NSF-funded S2S program focused on a 10-year analysis of a pair ofhigh-gradient modern systems (Keuhl et al., 2015), the Fly-StricklandRiver and offshore systems of Papua New Guinea, and the WaipaoaRiver S2S system in New Zealand (e.g., Driscoll and Nittrouer, 2000;Aalto et al., 2008; Francis et al., 2008; Jorry et al., 2008; Slingerlandet al., 2008; Carter et al., 2010; Marsaglia et al., 2010; Kuehl et al.,2015). The S2S results invited a closer look at ancient sedimentary sys-tems and, in particular, how quantitative analysis of a given part of thestratigraphic archive can be used to make inferences or predictionsabout upstream control, or downstream limits of strata. Such predic-tions are of keen interest to the hydrocarbon industry, especially inthe prediction of coarser-grained reservoir sediment volumes and qual-ity, prediction of deep-water reservoirs, and estimations of mud-pronehydrocarbon source-rocks and sealing facies (now also unconventional

reservoirs), which are key stratigraphic components of all petroleumsystems (Magoon and Dow, 1994). A key technique in characterizingand modeling subsurface hydrocarbon reservoirs is the use of bothmodern and outcrop analogs to constrain the size, shape, and complex-ity of sedimentary bodies (e.g., Reynolds, 1999; Bridge and Tye, 2000;Bhattacharya and Tye, 2004; Tye, 2004; Pranter et al., 2009; Deveugleet al., 2014). Picking a correctly scaled analog can be critical in reservoirprediction, and can be greatly informedby S2S concepts that help definethe nature of linked up-dip to down-dip depositional systems as a func-tion of the catchment parameters (Bhattacharya and Tye, 2004).

The focus of this paper is to review how S2S concepts are appliedto quantitative analysis of deep-time strata, with an emphasis onpaleodischarge and sediment budgets, to determine the scaling rela-tionships among the various components of the S2S system. The paperalso provides examples of how S2S concepts can be applied to petro-leum systems, especially in predicting sediment volumes, sandstone–shale ratios, and heterogeneity of conventional and unconventional res-ervoirs. Insights about sediment source areas, and especially climateand tectonics, provide predictions about reservoir quality, which islargely a function of sediment composition. The scale of a depositionalsystem is ultimately limited by factors such as catchment area, climate,and relief, which in turn control river discharge (Syvitski and Milliman,

Fig. 1. A. Source-to-sink components and processes. After Allen and Allen (2013) and Sømme et al. (2009). B. Channel networks in a typical S2S system. Tributary channels in the mainupland catchment area converge to form a larger trunk river, with smaller side tributaries. The trunk channel feeds a distributive fluvial system that may also form part of the coastalor delta plain. These can also be linked to deep-water systems, as shown in Fig. 1A.

79J.P. Bhattacharya et al. / Earth-Science Reviews 153 (2016) 77–110

2007) and the size and character of downstream linked depositionalsystems, such as deltas and submarine fans, and their components,such as lobes, channels, and bars.

A fundamental question is how S2S concepts can improve predictionof sediment partitioning among the alluvial, shoreline, shelf and deep-water environments, as embodied in the fulcrum approach ofHolbrook and Wanas (2014). This approach can improve prospectivityof reservoir volumes in different parts of a linked source-to-sink sedi-mentary system. This paper emphasizes examples from the MesozoicInterior Seaway of Western North America because they are very wellstudied; they contain prolific petroleum systems, and most exampleswere deposited in relatively closed systems, allowing sedimentarybudgets to be estimated.

2. Sequence-stratigraphy overview

Sequence stratigraphy provides a practical geometric approach tostratigraphic analysis which, when integrated with chronostratigraphicor biostratigraphic data, allows magnitudes of base-level change to bequantified (e.g., Posamentier et al., 1988; Nystuen, 1998; Catuneanu,2002; Bhattacharya, 2011)(Fig. 2). Sequence stratigraphy also providesa theoretical framework that explains how depositional sedimentarysystems are linked and how they evolve as a consequence of changesin accommodation. Initial sequence stratigraphic studies posited thatdepositional systems (such as rivers, deltas, and submarine fans;which also formkeyhydrocarbon reservoirs) showpredictable 3D strat-igraphic organization (Fig. 2), which were primarily related to cycles ofchanges in accommodation. Many submarine fans, for example, werehypothesized to largely form in the distal areas of sedimentary basinsduring lowstands of sea level (Posamentier et al., 1988; Van Wagoneret al., 1990; Nystuen, 1998). During sea-level highstands, more-distalenvironments were generally inferred to experience slower deposition,resulting in organic-rich condensed sections, whichmaymake excellenthydrocarbon source rocks, sealing facies, or unconventional reservoirs(Fig. 2).

In the original sequence stratigraphic models (e.g., Posamentieret al., 1988) sediment supply was assumed to be constant; lateral vari-ability was seldom considered, and most sequence stratigraphic sur-faces were considered to be chronostratigraphic. Since its originalformulation, there have been many modifications to the these models(e.g., Posamentier and Allen, 1993; Catuneanu, 2002; Strong andPaola, 2008; Catuneanu et al., 2009; Covault and Graham, 2010;Bhattacharya, 2011; Holbrook and Bhattacharya, 2012). There are nownumerous studies, especially of modern systems, which demonstratethat climate, sediment supply, and shelf width have a significant influ-ence on sequence formation and timing (e.g., Anderson et al., 2004;Walsh and Nittrouer, 2009; Covault and Graham, 2010). Sedimentmay be delivered directly to some deep-water systems during sea-level highstands, especially in settings where shelves are narrow orwhere sediment supply is very high (e.g., Walsh and Nittrouer, 2003;Covault et al., 2007; Carvajal et al., 2009; Covault and Graham, 2010).

Inasmuch as S2S recognizes boundaries between depositional sys-tems, the shoreline is fundamental in sequence-stratigraphic analysis.The concept of shoreline trajectory (Helland-Hansen and Martinsen,1996) involves tracking the migration of shoreline facies throughoutstratigraphic successions (Fig. 2). The stacking and trajectory ofshoreline deposits and their linked fluvial and offshore shelf and deep-water deposits, can be analyzed to estimatemagnitudes of accommoda-tion change (e.g., sea-level rise and fall), and potentially sedimentsupply (Neal and Abreu, 2009) and more recently sediment budgets(e.g., Carvajal and Steel, 2012; Guillocheau et al., 2012; Hampson et al.,2014). The shelf-slope break (Figs. 1 and 2) is also an important bound-ary that can be tracked inmany stratigraphic data; and is commonly ob-served in regional seismic lines. Cyclic alternations of strata have beenobserved in which sedimentary wedges that lie basinward of theshelf-slope break alternatewithmore extensive strata that extend land-ward of the shelf-slope break. The onlapping basinal wedges wereinterpreted as lowstand deposits, whereas the landward deposits havebeen interpreted as highstands (Vail et al., 1977; Fig. 2). Such regionalchanges in onlap patterns were used to reconstruct the history andmagnitude of sea-level change throughout geologic time (e.g., Vail

Fig. 2. A. Classic sequence-stratigraphy “slug” diagram. Lapout relationships, facies changes, and accommodation successions are used to interpret sequences and systems tracts. The ac-commodation successions also match shoreline trajectories of the delta/shoreface deposits. B. Time-stratigraphic relationships indicate main times of non-deposition. HST – Highstandsystems Tract, TST – Transgressive systems tract, SMST- Shelf Margin Systems Tract, sf –slope fans, bf –basin floor fans, mfs – maximum flooding surface, TS – Transgressive Surface,SB – Sequence Boundary. Modified after Vail (1987); accommodation succession concept from Neal and Abreu (2009).


et al., 1977; Haq et al., 1987), although not without controversy(e.g., Miall, 1997; Dickinson, 2003).

A fundamental aspect of the S2S perspective is that observations fromone part of the system, such as an incised river and its valley, potentiallypermit predictions to be made about other parts of the system, such asthe downstream shorelines or submarine fans fed by these valleys. This,of course assumes that sediment can be transferred across these bound-aries and that effects of downstream sea-level changes or upstream cli-mate changes can also be transmitted across these boundaries (Blumet al., 2013; Romans et al., 2015). Identificaion and correlation of keystratal-bounding surfaces permit time-equivalent depositional systemsto be correlated and mapped across a systems tract, which potentially al-lows consideration of sediment budgets across S2S boundaries(e.g., Hampson et al., 2014). As we increasingly understand the role ofsediment supply as a control on stratigraphy, it becomes more importantto quantify water and sediment fluxes in deep-time systems.

Other boundaries that are important landward of the shoreline in-clude the bayline and backwater limits (Fig. 3; Blum et al., 2013).).The bayline commonly marks the landward limit of marine facies andtidal deposits. The backwater length is the ratio of flow depth dividedby dimensionless river slope (Blum et al., 2013) and may control thegravel–sand boundary in some fluvial systems (Fig. 3) Backwaterlengths range from hundreds of kilometers in low gradientcontinental-scale systems, such as the Mississippi (flow depth is 40 mand slope is 5 × 10−5), down to a few kilometers in steep gradient sys-tems with shallow rivers (Blum et al., 2013).

There has been much recent work that focuses on the relative im-portance of upstream and downstream controls on sedimentary sys-tems (e.g., Blum and Törnqvist, 2000; Holbrook et al., 2006; Holbrook

and Bhattacharya, 2012; Blum et al., 2013). Sea level may be far less im-portant in controlling sediment accumulation and the behavior of riverslandward of the backwater length where climate and tectonics stronglyimpact discharge of sediment (Adams and Bhattacharya, 2005; Lawtonet al., 2015) (Fig. 4). In fact, these recent studies suggest that sea-levelrise or fall may have no effect landward of the backwater limit, indicat-ing that sea-level changes may not be the dominant control across theentire S2S transect. However, these concerns must be tempered withthe recognition that sediment produced by erosion of uplands is never-theless routinely transferred into basinal sinks. However, not all gradesof sediment are transferred downstreamof the backwater limit, and up-stream perturbations (i.e., signals) may be dampened or shredded dur-ing downstream transit (see Romans et al., 2015). There are numerousopen questions regarding how sediment transfer occurs. For examplewhat is the degree of intermediate staging of sediment along the trans-fer route, how do environmental signals propagate upstream anddownstream across key environmental boundaries, such as the shore-line and the backwater length,what are the time lags in sediment trans-fer across these boundaries, and what are the partition coefficientsbetween adjacent sinks (e.g., floodplain versus shelf), A number ofthese questions are addressed in the companion papers in this volume.

3. Critical considerations in applying S2S concepts to deep-timesystems

A critical aspect of deep-time systems is that over geological time(i.e., N106 years) much sediment may ultimately be transferred acrossan entire S2S system, even if temporarily stored in intermediatesedimentary sinks at shorter time scales (see also Romans et al.,

Fig. 3. Backwater and bayline length versus slope. A. cross section showing relative limits of brackish, tidal, and backwater limits. B. Slope versus backwater for a range of rivers (modifiedafter Blum et al., 2013).


2015). Conversely, a significant proportion of sediment can betemporarily or permanently stored in these intermediary sinks(e.g., the alluvial plain), and may subsequently or never be transportedacross an entire S2S transect. For example, there are numerous well-documented examples of extensive non-terminal fluvial and othernon-marine deposits, stored in fossilized intermediate sinks thatspan the rock record (e.g., Davies and Gibling, 2010; Gibling andDavies, 2012; Long, 2004, 2011; Ielpi and Ghinassi, 2015). Intermediatecontinental sediment storage can be important in low-gradient,continental-scale systems, and it is also important in the smaller-scaleCretaceous systems that we review in this paper. Syndepositional struc-tural features, such as growth faults and folds, can also accommodatesignificant volumes of sediment, especially on continental margins,and may preclude or reduce transfer of sediment to deep water, aslong as such faults are active (Galloway, 1986; Kuehl et al., 2015).These considerations affect how the stratigraphic record is used tomake inferences about sediment source areas, or downstream sediment

Fig. 4. Cross-section of highstand and lowstand fluvial and shelf deposits. The onlap limit of tlowstand to highstand. Depositional slope of the top of the lowstand may be measured basedto highstand. Also, elevation drop of the base of the fluvial long profile can be used to estimate

volumes, which may aid in prediction of down-dip sandstonereservoirs.

Another critical issue in deep-time systems is the chronometric res-olution of the deposition of strata (see also Romans et al., 2015) andtheir subsequent diagenesis or deformation. In the modern, sedimenttransfermay occur at a variety of time scales, fromdays up to a fewhun-dred years, but deep-time chronometers rarely have precision that isbetter than a fewhundred thousand years. In some cases, unique events,such as volcanic eruptions, large-magnitude turbidity flows,meteor im-pacts and the like, which deposit individual beds, can be tracked andcorrelated in deep-time records (Ricci-Lucchi, 1995; Bhattacharya,2011). Today's chronometric techniques can now routinely achieve anuncertainty of 0.5% of the total age, but this nevertheless can still repre-sent a large time span, e.g., half a million years for the Cretaceous rockswe are going to consider. A precision of 0.1%, which is becoming thenorm for modern geochronology, leaves an uncertainty of±94,000 years for strata whose estimated age is equivalent to the

he highstand coastal prism is a function of the slope versus vertical rise of sea-level fromon the onlap limit of the highstand prism and the change in sea-level rise from lowstandthe slope of the base of the lowstand incised valley (based on Blum and Törnqvist, 2000).


Cenomanian-Turonian boundary (93.9 Ma). The precision obtainedfrom isotopic methods will depend on the method used, and the meth-od used depends on the type of rock or sediemnt present (Dickin, 2005).Without appropriate lithology (e.g., a zircon-bearing rhyolite; bentonitelayer), the temporal resolution in a sedimentary sequence may belimited to understanding informed by the included fossils. As a conse-quence, linking short-term sedimentary processes, such as terraceformation in fluvial systems, to longer-term processes, such as theformation of sequence boundaries over the Milankovitch time-band,has long been problematic for deep -time studies (Sadler, 1981;Holbrook and Bhattacharya, 2012; Blum et al., 2013; Romans et al.,2015) But, it has been successfully achieved in Quaternary and someolder examples (e.g., Sømme et al., 2011, 2013; Sømme and Jackson,2013). Most recently, Singer et al. (2015) have achieved precision of0.02% using 40Ar/39Ar Multicollectors in Cretaceous systems,which promises the ability to resolve individual MilankovitchAstrochronological cycles.

Lastly, time-stratigraphic analysis of deep-time systems shows thatin any given vertical succession, very little geological time is actuallyrepresented by strata. Rather, many surfaces of non-deposition and ero-sion commonly separate stratal units (Barrell, 1917; Ager, 1973; Sadler,1981, Miall, 2015; Spencer et al., 2014 and references therein) (Fig. 2).However, this does notmean that the record is not preserved elsewherein the basin (Sadler, 1981). As a consequence, documentation of the fullstratigraphic record ideally requires 4D analysis, wherein Wheelerdiagrams add the 4th dimension of time (Qayyum et al., 2015). Such4D approaches of ancient S2S systems, have recently been attemptedin several cases (e.g., Guillocheau et al., 2012; Sømme et al., 2013;Hampson et al., 2014).

Quantitative S2S approaches also may use theoretical limits of sedi-ment production (e.g., BQART model of Syvitski and Milliman, 2007;Hinderer, 2012) to constrain the volumes of mud, sand, and gravelthat can be preserved in basins, as has recently been attempted byHolbrook and Wanas (2014) for Cretaceous strata in Egypt. This ap-proachwill be described for some of the Cretaceous examples discussedbelow. Prediction of sediment volumes is of fundamental importance tohydrocarbon resource estimates, and particularly conventional sand-gravel reservoir volumes, but may impact estimates of hydrocarbon-source rock volumes (which addresses volume of hydrocarbonresources) but this requires estimation of organic sequestration andconcentration. Leithold et al. (2015) specifically address carbon cyclingand organic sequestration in a variety of S2S systems, and this is criticalto predicting petroleum systems, especially source-rock quality and un-conventional shale reservoirs.

Fig. 5.River discharge versus drainage area. FromMulder and Syvitski, 1995 after Matthai,1990.

4. Components of S2S dispersal and depositional systems

The building blocks of all S2S systems include the catchment(source) area, a transfer system (usually rivers), and a series of linkeddown-dip depositional systems, including shoreline, shelf and deep-water systems (Fig. 1). The first-order control on the ultimate size ofany given S2S system is the size of the catchment area available to pro-duce sediment (e.g., Matthai, 1990). To a lesser degree, the relief and cli-mate of the catchment area play a role (Figs. 5 and 6). The area and reliefof the catchment can be measured directly in modern systems(e.g., Syvitski and Milliman, 2007), but are typically more difficult toestimate in deep-time systems. Reconstructing the source area mayrequire reconstruction of drainage basins based on drainage dividesand plate-tectonic reconstructions (Fig. 7), especially in areas wheretectonics have deformed or removed the source area, such as commonlyoccurs along rifted margins. Despite this limitation, there are a numberof approaches to estimating the area, relief, and sediment yield from anancient system (e.g., Bhattacharya and Tye, 2004; Miller et al., 2013;Holbrook and Wanas, 2014; Carvajal and Steel, 2012). In addition,source areas may be identified by provenance analysis, such as the dat-ing of detrital zircons (e.g., Dickinson and Gehrels, 2008; Miller et al.,2013; Blum and Pecha, 2014) (Fig. 7), as discussed in Section 6.7.

Ideally, estimating S2S parameters should take a statistical or proba-bilistic approach in which a range of values may be estimated to placeerror bars on possible maximum and minimum sediment volumesand consequently, the scaling of linked depositional systems along aS2S transect. Particularly useful are scaling relationships, such as recent-ly explored by Sømme et al. (2009), who demonstrated correlationsamong a number of measurable parameters based on a compilation ofQuaternary S2S systems. They showed positive correlations betweenriver-length and submarine fan length, and an increase in submarinefan area with increasing catchment areas (e.g., Fig. 6), and a negativecorrelation between catchment area and shelf gradient. The scalingrelationships are typically empirical in nature and involve searchingfor relationships (e.g., power laws) that demonstrate how parameterssuch as catchment area and river discharge are related.

4.1. Alluvial systems, catchments and transfer

In this paper we treat the catchment (i.e., drainage basin) as an areaof net sediment production (i.e., export), especially where there is net

Fig. 6. “Conveyor-belt” vs. “vacuum-cleaner” models for sediment supply. Conveyer-beltmodel indicate long-lived sediment supply from a large hinterland drainage basin in apiedmont valley, vs. local excavation of a coastal-plain paleovalley. The conveyer-beltmodel represents an extra-basinal river system, draining a larger catchment area, whichshould build a larger fan, compared to the smaller-scale, intra-basinal vacuum-cleanermodel.Modified from Blum and Törnqvist (2000) and Blum et al. (2013).

Fig. 7.A. Paleogeographic reconstruction of early Triassic circum-Arctic drainage areas based on B. Detrital zircon analysis of Triassic outcrops around the circum-Arctic Ocean. Vertical axisin B refers to thepercentage of age-specific detrital zircons versus the total populationmeasured. Variations in age-spectra are associatedwith age of source areas. FromMiller et al. (2013).


uplift. However, we recognize that the catchment area may include itsown depositional elements, such as alluvial fans and talus deposits(Fig. 1). For the purpose of this paper, we focus on rivers as the primarytransfermechanism for sediment. See Jaeger et al. (2015) for insights oncryogenically controlled S2S systems. Rivers often show a tributary pat-tern in the catchment area; channels join downstream to form a singletrunk stream in the middle reaches (Fig. 1). On the flatter coastalplain, rivers commonly form a distributive pattern, creating either anaggradational distributive fluvial system (e.g., Weissmann et al., 2010)or a deltaic deposit. The length of the medial to distal parts of the S2Ssystem (i.e., the area between a mountain front and the shoreline),

Fig. 8. A. Campanian versus B. Paleocene paleo-drainage patterns in North America. During theresulting in numerous separate small delta systems and associated local drainage areas. Durinbecome integrated and coalesce to form much larger S2S systems that feed deep-water fans,and larger areas of land are exposed. Green outlines of paleo-land-masses largely based on recbased on Blum and Pecha (2014). Gulf of Mexico Paleogeography from McDonnell et al. (2008

and the area and relief of the potential catchment area, control thescale of the largest trunk rivers (Figs. 5 and 6) and this appears to affectthe size of deltas and submarine fans (Sømme et al., 2009).

Blum et al. (2013) point out that continental-scale, integrateddrainage basins, such as the Mississippi, are more likely to develop inlow-gradient settings during icehouse times (e.g., Fig. 8). Sea-levelfalls across a low-gradient shelf during an icehouse, expose large areasresulting in the development of a wide coastal and alluvial plain andthus gives more opportunity for development of integrated drainagesbasins at times of low sea level. During Greenhouse times, higher sealevels, and consequent flooding of continents, created smaller, less

Cretaceous “Greenhouse”when global sea-levels were high, (A), the continent is floodedg up-lift of North America during the later Laramide orogeny (B), the smaller drainagessuch as in the Gulf of Mexico. This is also analogous to Icehouse times when sea retreatonstructions of R. Blakey, and Colorado Plateau Geosystems (2015). Paleodrainage in B is).


integrated drainage basins resulting in smaller and more isolated S2Ssystems, such as those that characterized much of the Cretaceous ofNorth America (Fig. 8). Of course, tectonics also play a major role incatchment development and scale, such as during early break of Pangeawhen continents were at higher elevations or during major orogenicplateau development.

For any given sedimentary basin, the contributing rivers may draindifferent areas. These fluvial systems can range from the largest-scaleextra-basinal rivers that are linked to bedrock catchment areas tosmaller-scale basin-fringe rivers, or the smallest-scale intra-basinalstreams lying entirely within a depositional basin (Galloway, 1982).Blum and Törnqvist (2000), pointed out that extra-basinal river systemsact as conveyer belts, transferring sediment from upland catchmentareas across and into the depositional basin, whereas intra-basinal val-leys simply provide local material to more distal downstream sinksforming smaller-scale depositional systems (Fig. 6). Extra-basinal sys-tems have also been described as piedmont incised-valley systems(Boyd et al., 2006) in which a “fall line” typically marks the transitionfrom steeper to lower gradients. Downstream of the “fall line” both de-position and transfer of sediment can occur, whereas upstream erosiondominates (Figs. 1 and 6). Piedmont-valley systems are distinguishedfrom coastal-plain valleys, in that the latter occupy the lower deposi-tional reaches of a S2S system (Fig. 2). This paper will primarily focuson extra-basinal, piedmont-valley systems, as the associated trunkstreams contain information about the entire upstream source area,and can be used to determine the area, relief, and provenance, but weemphasize here that it is important to attempt to distinguish thesevalley types in ancient systems.

The total sediment load, or volume, associated with a S2S system isprimarily related to the size of the drainage area, although relief andclimate play a secondary role (Matthai, 1990; Milliman and Syvitski,1992), whereas the sediment yield (i.e., production of sediment perunit area) is more sensitive to climate and relief. This is critical indetermining which type of S2S systems are more effective in deliveringsediment to downstream sinks. Sediment yield plays a big role in thenature of oceanic delivery processes, such as hypopycnal versushyperpycnal flows (Milliman and Syvitski, 1992; Mulder et al., 2003).Estimates of sediment yield and sediment load, integrated with thescale of the S2S system, can beused to predict thenature of downstreamfacies, with implications for unconventional oil and gas resources.

4.2. Shorelines, shelves, and along- and across-shelf transfer of sediment

A significant sink formuch riverine sediment is the shoreline and ad-jacent shelf (Fig. 1). There are a variety of sub-environments withinthese depositional systems, including subaerial deltas, shorefaces,estuaries and their coeval offshore muddy shelves, typically manifestedas mid-shelf mud belts (Hill et al., 2007). Sedimentation may occur rel-atively close to the river mouth (Walsh and Nittrouer, 2009), such as ina river-dominated delta system, and there is increasing recognition thatsmall rivers that drain mountainous areas can deliver sediment directlyonto the shelf via hyperpycnal flowswhen they experience flood condi-tions (Mulder and Syvitski, 1995; Parsons et al., 2001; Mulder et al.,2003). Sediment gravity flows, (such as turbidity currents, fluid muds,and hyperpycnal flows), are fundamental mechanisms that transfersediment to shelf and deep water, and S2S analysis can help predictwhich types and scales of river systems generate these types of flows.In addition, a variety of shelf processes (chieflywaves, tides, and oceaniccurrents) can transport significant sediment far from the point of fluvialentry to seaward areas (Hill et al., 2007; Walsh and Nittrouer, 2009).Sandy bedload sediment may be moved and deposited alongshore aswave-dominated shoreface deposits (Plint, 2010; Bhattacharyaand Giosan, 2003). Mud can also be transported along-shore to forma chenier plain (Warne et al., 2002; Rine and Ginsburg, 1985;Bhattacharya, 2010; Bentley et al., in this volume) or may be deposited

farther offshore to form a mid-shelf mud belt (Fig. 9) (Walsh andNittrouer, 2009 and references therein).

There have been a recent spate of studies documenting inner-to mid-shelf mud belts in which exceedingly large volumes of fine-grained sediment (chiefly clay and silt) may be transported over dis-tances thatmay exceed a thousand kilometers from the point of riverineor deltaic depositional input (Hill et al., 2007; Liu et al., 2009; Walshet al., 2015). In the 1980's, it was thought that mid-shelf geostrophicflows, typically linked to storms, moved significant quantities of sandand gravel (e.g., Bergman and Snedden, 1999), but these models werelater revised with the recognition that so-called “shelf” sands are morelikely to be deposited during sea-level falls and left behind as top-truncated shoreface and delta deposits resulting in basin-isolatedsandstones (Plint, 1988; Bhattacharya and Willis, 2001; Martinsen,2003a, 2003b). It is now clear that incremental geostrophic transportis an important process for sediment transport on marine shelves(e.g., Suter, 2006; Bentley, 2003; Hill et al., 2007), but it is primarilyresponsible for the transport of mud rather than sand or gravel. Thisalong-shelf transport must be considered in S2S budget analyses.

Other breakthroughs in research on mud transport include thepioneering work of Schieber et al. (2007), who suggested that bed-load transport of floccule-mud ripples may be far more important inshelf systems than previously thought. These ideas are now being ap-plied to the re-interpretation of ancient mudstones, previously thoughtprimarily to reflect passive settling by suspension from hypopycnalflows during times of quiescence in the associated sea or lake sink(Bhattacharya and MacEachern, 2009; Plint et al., 2009). Wave-supported sediment gravity flows have also been related to ancientstrata (Macquaker et al., 2010).

This array of processes have enormous implications for S2S systems,because of the potential extent of along-strike and across-shelf sedi-ment transport and the need to account for this “export” in sedimentbudget studies (Draut et al., 2005; Hill et al., 2007; Kuehl et al., 2015).This is also important from the perspective of sediment budgets. Inmost S2S systems 70–95% of all the sediment carried by rivers com-prises fine-grained to muddy suspended load, and accounting for all ofthis mud can be challenging in both modern and ancient systems(Walsh and Nittrouer, 2009; Kuehl et al., 2015). With the rise in impor-tance of unconventional reservoirs, and especially shale plays, there hasbeen renewed economic interest in the sedimentology and stratigraphyof mudstones, and this will be discussed with respect to several of theCretaceous examples in Section 7.

4.3. Deep-water systems, the ultimate sink

Deep-water systems, such as submarine fans, are often considered tobe the ultimate sink (Figs. 1, 2, 6 and 8). Sediment delivery to deepwater is favored during lowstand conditions, when rivers debouch atthe shelf edge (Mallarino et al., 2006; Posamentier and Walker, 2006).However, large river-deltas may reach the shelf edge or connect to can-yons during highstand times, especially if the shelf is narrow or wheresubmarine canyons tap into the littoral drift system, such as occurstoday along the California coast (e.g., Anderson et al., 2004; Walsh andNittrouer, 2003, 2009; Covault et al., 2007; Covault and Graham, 2010;Romans et al., 2015). Volumes of sediment in submarine fans can belinked and compared to long-term river discharge and have been usedto estimate sediment budgets and sediment partitioning in Quaternaryand older systems (e.g., Goodbred and Kuehl, 1999, 2000; Carvajal andSteel, 2012; Petter et al., 2013; Sømme et al., 2011, 2013; Sømme andJackson, 2013).

5. Source area versus discharge

A significant aspect of the S2S concept is linking sediment sourceareas to the depositional record. In many modern S2S systems, muchof thewater (and sediment) is routed through a small number of incised

Fig. 9.Orinoco andAmazon S2S systems. 50% of themud in theOrinoco delta is derived from theAmazon rivermouth (dark gray), representing 1100kmof along-shelf transport, primarilyby wind-driven waves (Warne et al., 2002). Mud is also sequestered along the Suriname Chenier plain and inner shelf (brown color) (Rine and Ginsberg1985).


trunk rivers. Discharge measurements from these rivers can in turn belinked to the area, relief, and climate conditions of the drainage basin.For example,Matthai (1990) defined empirical power-law relationshipsthat showed how drainage area could be used to predict discharge ofdownstream trunk rivers (Fig. 5), although these discharge estimatesvary over an order of magnitude for any given area.

Syvitski andMilliman (2007) recently introduced themore sophisti-cated empirical BQART equation, based on analysis of hundreds ofmodern systems:

Qs ¼ ωBQw0:31A0:5RT

In this formulation, sediment discharge (Qs, 106t/yr) is related towater discharge (Qw, km3/yr), basin area (A, km2), relief (R, km), and av-erage temperature (T, in C°). The B term accounts for lithology, glacialerosion, and anthropogenic effects (see discussion in Syvitski andMilliman, 2007), but glacial erosion and anthropogenic effects did notoccur in the Cretaceous systems that we discuss below and can thusbe ignored. The equationwas developed for environmentswith averagecatchment temperatures greater than 2 °C, where ω = 0.0006. Sømmeet al. (2009) used this approach to estimate S2S budgets in the Quater-nary Golo River in Corsica, France, and its linked submarine fans, andshowed significant improvements in the uncertainty of estimates by in-corporating relief and temperature versus just catchment area. Carvajaland Steel (2012) used this equation inversely to estimate relief in thelate Cretaceous Washakie basin in Wyoming, given that they hadsome control on the long-term sediment discharge from analysis ofclinoform stacking patterns.

Paleogeographic and plate tectonic reconstructions of an ancientsystemmay be used to characeterize the sediment source area, espe-cially when linked with provenance data, such as detrital zircons andother heavy minerals (e.g., Dickinson and Gehrels, 2008; Miller et al.,2013; Blum and Pecha, 2014). Relief may be estimated fromthermochronometry, landward projection of slope estimates, or

from geodynamic modeling (Allen and Allen, 2013), and tempera-ture may be estimated from climate proxies (e.g., paleosols).

In many sedimentary basins, paleorivers drained the same areasas overlying modern rivers, allowing easier estimations ofpaleodischarge. The Mississippi drainage basin, for example, hasbeen supplying sediment to the Gulf of Mexico for much of the Ceno-zoic (Winker, 1982; Galloway et al., 2011; Blum and Pecha, 2014,Bentley et al., in this volume; Fig. 8). In other systems, paleoriversand their associated drainage basins may be separated by rifting orother tectonic processes. For example, the source area that suppliedthe Permo-Triassic sandstones of the Sadlerochit Formation in theBrooks Range, Alaska, was rifted away from the depositional basinduring the Jurassic opening of the Canada basin (Grantz et al.,1990), making it difficult to link the source to its sink (Fig. 7). TheSadlerochit Formation now lies on the Arctic–Alaska microplate,and interpretations about the original source area must be inferredfrom stratigraphic analysis, facies distributions, paleocurrent analy-sis, plate tectonic reconstruction, and provenance analysis, includingdetrital zircons (e.g., Miller et al., 2013, Fig. 7). Linking source andsink in this case have important resource implications because far-ther to the northwest, the Sadlerochit Formation contains the IvishakSandstone, which hosts the supergiant, 24 billion-barrel PrudhoeBay Field (Tye et al., 1999; Bhattacharya and Tye, 2004).

Rivers can also change course, as a consequence of local tectonics. Forexample, Wernicke (2011) suggested that during the Late Cretaceous,the paleo-Colorado River originally drained California, cutting the initialphase of the Grand Canyon and flowing east. Later Laramide upliftchanged the direction of the Colorado River to flow southwest andfinishing the carving of the modern Grand Canyon. A good example ofriver-course changes in older strata can be found in the Campanian fluvi-al deposits in the western interior of the USA, where renewed upliftcaused significant paleogeographic reorganization and provenancechange of rivers (e.g., Lawton et al., 2003; Adams and Bhattacharya,2005).


6. Paleodischarge, paleoslope, sediment budgets, and source-areaestimates in deep-time systems

There have been a number of attempts to constrain S2S budgets indeep-time systems using a variety of approaches. One approach is toestimate both water and sediment paleodischarge using outcrop andsubsurface data (Bhattacharya and Tye, 2004; Davidson and North,2009; Li et al., 2010; Musial et al., 2012; Holbrook and Wanas, 2014).Water discharge (Qw) of modern rivers varies over about 6 orders ofmagnitude (1 m3/s up to 105 m3/s), and comparison with modern ana-logs is a key component in testing predictions. The BQART method ofSyvitski and Milliman (2007) has also been applied to Quaternary sys-tems (e.g., Sømme et al., 2009) although this requires informationabout the relief and area of the catchment basin. Sediment budgetsand partitioning can also be calculated from chronostratigraphicallyconstrained stratigraphic studies, in which the degree of aggradationversus progradation of clinothems is linked to sedimentation rates andcan be compared with mapped volumes of sediment in distal shelfand/or submarine fan sinks (e.g., Hampson et al., 2014; Carvajal et al.,2009). Petter et al. (2013) and Gong et al. (2015) estimate sedimentflux in ancient systems using 2D analysis of clinoforms. The generalapproach to calculating paleodischarge from ancient river deposits isoutlined below followed by some specific examples.

6.1. Characterizing the trunk river

The first step in estimating paleodischarge of a S2S system is to findthe largest trunk river in a given stratigraphic unit (e.g., formation,member, or sequence stratigraphic subdivision). The largest trunk riverstypically lie within incised valleys above regional sequence boundaries(Boyd et al., 2006) (Figs. 1 and 10). At lowstands, exposed land areasare typically maximized, and thus drainage basins are most fully devel-oped. The ability to identify the largest rivers in any particular formationis favored by areas where extensive outcrops are available (Figs. 10and 11). For example, the southwestern deserts of the U.S.A. providestens to hundreds of kilometers of cliff exposures of Mesozoic forelandbasin fluvial successions (e.g., Van Wagoner and Bertram, 1995 and

Fig. 10. Erosional scour marks a base-of-valley sequence boundary (SB) overlain by a floodpchannel (about 5 m deep and 20 m wide) from the larger valley-scour, Ferron Notom delta sy

references therein; Garrison and van den Bergh, 2004; Pranter et al.,2009; Li et al., 2010; Hajek and Heller, 2012; Hampson et al., 2013).

Valley systems may also be identified and mapped where there isextensive subsurface data such as extensive well log data bases(e.g., Plint, 2000, 2002; Plint and Wadsworth, 2003, 2006; Carvajaland Steel, 2012). Well logs may penetrate channel fills and provide de-tails of channel depth. 3D seismic data may image river deposits(e.g., Musial et al., 2012; Reijenstein et al., 2011; Miall, 2014) and, de-pending on seismic resolution, width and depth of channels may be ex-tracted (Fig. 11).

6.2. Estimating channel dimensions and sediment characteristics

Key channel dimensions, including bankfull depth, width, and cross-sectional area may be measured or estimated. Grain size, sorting, andsedimentary structures (Fig. 12) can be used to quantify bed shear stressand to interpret bedforms and flow regime. Channel depth may be esti-mated on the basis of the thickness of fully preserved fining-upward fa-cies successions, which represent individual channel stories, as seen inoutcrop (Figs. 2, 10 and 12), cores or well logs. Maximum bar heighttypically represents about 80% of flow depth. Fully preserved bars maybe marked by finer-grained drapes or a distinct rollover in accretionbedding, both of which may indicates the bar top (e.g., Hajek andHeller, 2012; Blum et al., 2013; Wu et al., 2015; Ullah et al., 2015,Holbrook and Wanas, 2014). Flow depths can also be calculated usingthe average thickness of dune-scale cross-sets (LeClair and Bridge,2001; Adams and Bhattacharya, 2005; Li et al., 2010). Channel widthcan be calculated based on empirical equations (see references inBridge, 2003), or may be estimated in well-exposed outcrops (Figs. 10and 11C), such as the flow-perpendicular cliff exposures that can com-monly be found in the outcrops of the western interior of NorthAmerica. In caseswhere point bars arewell exposed, thewidth of lateralaccretion surfaces typically represents 50% to 80% of channel width(Bridge, 2003; Donselaar and Overeem, 2008). Channel width can alsobe estimated using plan-view images from 3D seismic data sets (e.g.,Reijenstein et al., 2011; Musial et al., 2012; Miall, 2014), and rarely,plan view exposures in outcrops (Ielpi and Ghinassi, 2014; Wu et al.,2015) (Fig. 11C). Holbrook and Wanas (2014) suggest that where

lain and its meanderbelt. Teal-colored dotted line distinguishes the smaller meanderingstem, Utah.

Fig. 11. A. Seismic time slice and block-diagram showing incised meander-belt (Gulf of Thailand). Mud-filled channel is about 300 m wide. Cross-section view shows laterally-accretingmeander-belt (from Reijenstein et al., 2011). C. Plan-view exposure of meander-belt in the Ferron sandstone with paleochannel reconstruction (FromWu et al., 2015). Channel is about50 m wide.

Fig. 12.Measured section through fluvial deposits of the Ferron SandstoneMember, Utah(Li et al., 2010; Zhu et al., 2012). Fining-upward facies successions (arrowed lines) markchannel stories. Thickness of cross-sets and fining-upward channel stories can be usedto estimate flow depths. Thickest stories in a formation can represent trunk rivers.


cross sectional or plan-view data are absent, empirical width estimatesmay be accurate to a factor of 2–4 (Fig. 13), but when present widthestimate errors are considerably lower (b2), but need to consider theorientation of the exposure with respect to paleocurrents.

Bedload sediment is easily measured in the field with a standardgrain size card, usually providing a 1/2-phi precision. Sorting canalso be estimated visually. Bedforms are interpreted based on obser-vation of sedimentary structures, such as cross bedding, which isformed by subaqueous dunes, or cross lamination, which is formedby ripples. Thickness of dune-scale cross sets should also be com-piled as they can be used to calculate flow depths (Leclair andBridge, 2001). Paleocurrent data is used to determine the orientationof the outcrop face with respect to flow direction, which is critical forinterpreting bar and channel types (i.e., braided versus meandering,Bridge, 2003), and simple trigonometry can be used to adjust appar-ent widths to true widths, especially for morphological reconstruc-tions of channel aspects from point-bars.

Measurements are typically compiled in standard measuredsections, showing thickness, grain size, and sedimentary structures(Fig. 12). The depth of channels is relatively easy to measure in thefield, with a precision of about 1 m using standard tools(e.g., Jacob's staff or measuring tape). Channel and bar dimensionscan be obtained from facies architectural analysis based on beddingdiagrams of photomosiacs (Fig. 14). Lidar and photorealistic datacan be used to avoid parallax, allowingmore accurate measurementsof channel and bar geometries (e.g., Pranter et al., 2009; Olariu et al.,2011; Hajek and Heller, 2012). Errors associated with field measure-ments, especially grain size and thickness are low, typically on theorder of b10% (Fig. 13). Errors in width estimations are higher (typ-ically 50–100%), and depend on the quality of exposures (Fig. 13). Er-rors in empirical equations that relate width and depth(e.g., Williams, 1988) also have an error of about a factor of 2(Fig. 13). Width estimates are much less certain in braided streamsthan in single-thread meandering rivers, and the distinction be-tween braided and sinuous channels is not always easy to make, es-pecially in subsurface.

Fig. 13. Tornado diagram showing range of errors in estimation of paleodischarge and sediment budgets using the fulcrum method, from Holbrook and Wanas (2014).


6.3. Paleoslope estimates

Slope is a fundamental control on S2S systems. Alluvial fans typicallyexhibit slopes in excess of 1.5° (Blair and McPherson, 1994) whereasnatural river gradients are much lower and range over about 3 ordersof magnitude (10−3–10−5) (Fig. 3). Slope controls many aspects ofthe morphology of rivers and also influences the position of specificS2S boundaries, such as the shoreline, backwater, and bayline (Figs. 3and 4). For example, slope affects river plan form (e.g., braiding isfavored by higher slopes and meandering by lower slopes), depth ofriver incision, and facies boundaries, such as the gravel–sand and limitof tidal influence (Bridge, 2003; Blum et al., 2013).

Fig. 14. Photomosaic,measured sections, bedding and facies diagramofmeander-belt sandstonelamina-sets (Sr) are not shown in the bedding diagram, but can be seen in the measured sectioversus subaqueous dunes. Thickness of fill approximates flow depth, and width of lateral accre

Measuring low slopes in modern systems is easy; one simply calcu-lates the elevation drop over somedistance; however, measuring slopesin ancient systems is more challenging. Slope may be calculated usingphysics-based methods, such as the Shield's thresholds for bedloadmovement, especially where the grain size, flow depth, and width of apaleo-river are known (Bhattacharya and Tye, 2004, Bhattacharya andMacEachern, 2009, Davidson and Hartley, 2010; Musial et al., 2012;Holbrook and Wanas, 2014), as described in Section 6.4.

Blum and Törnqvist (2000) suggest a simple geometric method forestimating paleoslope based on onlap limits of river-fed clastic wedgesformed as a consequence of sea-level changes (Fig. 4). An alternatestratigraphic approach measures the elevation drop of coeval channel

in the Ferron Sandstone. FromWu et al., 2015. Dune-scale cross beds (St) and ripple-crossn and facies cross section. Large cross-sets (Sf) are associated with laterally accreting barstion beds are about 80% of flow width. Channel belt erodes into floodplain deposits (Fm).

Fig. 15. A. Depositional dip cross section of the Ferron Notom delta, Utah. From Zhu et al. (2012). B. Slope calculations from stratigraphic cross section. Ferron slopes are quite steep:7.6 × 10−4 – 2.6 × 10−3 (0.00076–0.0026) or 0.04° – 0.14°.


or valley floors (i.e., parts of the fluvial long profile), with respect toa flat stratigraphic datum in flow-parallel cross-sections (Figs. 4and 15). Estimates of slope, obtained from such stratigraphic data, canbe used to estimate flow velocity and paleodischarge and provide anindependent estimate that can then be compared to those made usingthe methods based on grain size, bedforms, and channel depths, asdescribed below.

The errors in these geometric-stratigraphic slope estimates dependon the precision and accuracy of measurements of the thickness andlength of onlapping clastic wedges. The precision of length and thick-ness measurements in stratigraphic studies using standard measuringtools (e.g., Jacob's-staffs) depends on the scale of measurement, butare typically about 10% (e.g., thickness of 10 m ± 1 m, and lengths of1 km ± 100 m). The accuracy of the measurement also is related tothe orientation (i.e., obliqueness) of the cross section, especially iftaken from 2D cross sections. Although the thickness of the wedge isnot sensitive to orientation of the cross section, the horizontal extentof the wedge is strongly dependent. Longer apparent widths yield erro-neously low slope estimates. Error may be resolved if the wedge can bemapped in 3D, and thus measurements can be made perpendicular tothe shoreline. Also, if the degree of obliqueness can be determined, ageometric correction can made using simple trigonometry. Compaction

can also effect thickness estimates andmay result in underestimation ofslope. Compaction of about 50% can cause underestimation of slope by afactor of about 2 or more, but probably not an order of magnitude. Theuse of these stratigraphic techniques yields slope estimates within anorder of magnitude, and perhaps better in high-quality datasets.Geometric-stratigraphic methods can also be compared with physics-based estimates, as described below.

6.4. Velocity and paleodischarge estimates

Flow velocity may be estimated by plotting grain size, bedform, andflow depth of a fluvial deposit on the 3D bedform phase diagram ofRubin and McCulloch (1980) (Fig. 16). This yields an estimate that typ-ically ranges over a factor of 2. Alternatively Law-of-the-Wall relation-ships, or estimates based on Shields criteria for grain movement canbe made based on grain size and sorting measurements (Musial et al.,2012; Holbrook and Wanas, 2014). For such calculations, bedloadgrain-size distribution (e.g., D50) is required. The average velocity (U)is multiplied by the estimated channel cross-sectional area to obtain adischarge estimate (Qw). If slope is known, the Manning or Brownlie(1983) equations can also be used to estimate velocity and discharge,

Fig. 16.Velocity estimates for the Dunvegan and Ferron trunk channels forfine tomediumsandstone, water depths of about 5 m to 20 m, and with observations of dune-scalebedforms. Velocities of about 0.6 to 1.8 m/s are estimated based bedform phase diagramsof Rubin and McCulloch, 1980.


respectively, for a river of known depth, slope, and roughnesscoefficient:

Manningequation;U ¼ 1=n Rh0:66 S0:5

WhereU is average velocity, n is theManning coefficient (in s/m0.33);Rh is the hydraulic radius (m), and S is the dimensionless slope.

Table 1Paleohydraulic estimates for a variety of Cretaceous Formations discussed in the text.

River Bankfullthalwegdepth (m)

Meanbankfulldepth (m)

Width(m)

Slope U m/s Water

Qw 103m

Mississippi 40.0 – 1500 0.00005 1.0 15Lance 7 – – – – –Blackhawk 9.8 7.2 – 0.0007 – 0.4–2.2Ferron V1 8.4 4 140 0.0008 0.75–2.3 1.3–3Ferron V2 5.5 3 70 0.0007 1.9 0.4Dunvegan 16–20 11 170–200 0.0003–.00005 0.75–1.6 1.6–6Paddy 7.0 – 100 – – –McMurray 40 – 800 0.0003 0.75–1.5 15–34Paddy 7.0 – 100 – – –Chinle/Dockum 10–15 – – – – –

The Brownlie (1983) equations can also be used:

Q0:6539 ¼ R=0:3724S−0:2542σ s−0:1050 D50

whereQ ¼ Qp=Wg0:5D501:5

wherein the dimensionless discharge (Q), is calculated based on the hy-draulic radius (R), channel gradient (S), grain sorting (σs) mean grainsize (D50), paleodischarge (Qp), channel width (W) and gravitationalconstant (g). Where river depth and velocity are already known, theseequations can be inverted to calculate slope. Table 1 shows estimationsfor a number of examples.

Musial et al. (2012) used the Brownlie (1983) equation to estimatepaleodischarge of the river deposits of the Cretaceous McMurray For-mation in Canada (Fig. 17). They had data on grain size, point-bar thick-ness (30 m) and channel width (750 m), and estimated slopes of 10−4.Using these numbers they calculated a Qw of 15 × 103 m3/s, about thesame as themodernMississippi River. We have independently estimat-ed the velocity using the 3D bedform diagram approach describedabove. The 40 m deep McMurray channel comprise dune-scale crossbedded, medium-grained sands. Plotting these on the bedform phasediagram yields a bankfull velocity estimate of between 0.75 m/s and1.5 m/s. The corresponding Qw ranges from 17 × 103 m3/s to34 × 103 m3/s, concurring with the Qw calculations of Musial et al.(2012).

Velocity estimates using theManningor Brownlie equations are sen-sitive to uncertainties in slope estimates and grain-size distribution.Grain size of the bedload fraction can be estimated quite precisely, butslope estimates can typically bemade only to within an order of magni-tude (see below). Although there is little uncertainty inmeasuring grainsize and determining bedform type, both bedforms and grain size canvary considerably in any given channel deposit, reflecting both verticaland horizontal variations in flow. Maximum velocities, however, willbe reflected in bedforms moving in channel thalwegs and transportingthe coarsest grain sizes as bedload (rather than lag deposits). Ideally, ve-locities should be estimated using both the Manning and Brownlieequations as well as the bedform phase-diagram to test for congruityof values. Velocity estimates made using the bedform-phase diagram(which does not incorporate slope as a dependent variable) can alsobe used to estimate slopes.

6.5. Paleodischarge estimates using regional hydraulic geometry curves

Davidson and North (2009) use regional hydraulic geometry curvesto estimate the discharge and drainage area of a number of ancient riversystems based on the drainage area. These relationships are developedfrom analysis of rivers from different climatic and tectonic areas. Appli-cation to an ancient system requires someknowledge of the climate andmorphological regime of the ancient drainage area (e.g., arid or humid,temperate or tropical, high or low latitude, mountainous or lowland).

Sediment DA103 km2

Fulcrum calculationof sed. volume in20,000 years km3

Sandstoneisopachvolume km3

Total Sedimentisopach volumekm3

3/s Q b m3/s Q t m3/s

– – 320 – – 300,000– – – – 66 630– – – – – –0.70 11.2 50 17.60 20 2000.33 2.8 50 0.83 5 50– – 125 – 20–75 3000–7000– – – – – –– – ~500 – 300 30,000– – – – – –– – 6–200 – – –

Fig. 17. Point Bars of the McMurray, Formation, Alberta, Canada. A. seismic time slice taken from about 400m below the surface. B. Interpretation of seismic time slice showing point barsand abandoned channels, C. Three-dimensional reconstruction showing paleochannel and paleohydraulic estimates. Inclined point-bar lateral accretion surfaces are locally draped byshale forming inclined heterolithic strata (IHS) that inhibit fluid flow and contribute to the reservoir heterogeneity. From Musial et al. (2012).


Such climate informationmay be derived from direct observation of thesedimentary facies (e.g., analysis of floodplain paleosols) or plate-tectonic and paleogeographic reconstructions, which place a systemwithin a corresponding paleo-climate belt. The data for this approach

Fig. 18. Plot of drainage area versus bankfull discharge for climate regimes in Florida. FomDavidson and North (2009).

require estimates of bankfull channel depth, which can be estimatedfrom point-bar dimension, or channel-fill thickness, as well inferredclimatic and tectonic regime. Regional hydraulic geometry curves(Fig. 18) use power-law relationships of the form:

D ¼ aAb

orQw ¼ aAb

(Davidson and North, 2009), where D = channel depth, Qw =discharge, andA=drainage area. The constants, a and b, are empiricallydetermined using survey and streamgauge data fromanalog catchmentareas (Davidson and North, 2009 and references therein); one exampleis shown in Fig. 18. Qw and D may be estimated from the field orsubsurface using the approaches described above. Errors and uncer-tainties associatedwith use of these curves typically ranges over a factorof 2–4, and up to an order-of-magnitude.

6.6. Estimates of sediment budgets from paleodischarge

Expanding on the techniques described above, Holbrook andWanas (2014) calculated longer-term sedimentation rates to esti-mate sediment budgets of Cretaceous paleo-rivers, exposed inEgyptian outcrops. This required estimation of annual sediment dis-charge, which in turn required estimating flood frequency. The bulkof sediment transfer in modern rivers typically occurs during large

Fig. 19. Paleodrainage of river systems developed in the Triassic. Compilation based onGibson and Barclay (1989) for the Alberta systems; Dubiel (1994) and Dickinson andGehrels (2008) for the Chinle-Dockum, and Miller et al. (2013) for Arctic Canada. Greenoutlines of paleo-land-masses largely based on reconstructions of R. Blakey and ColoradoPlateau Geosystems (2015).


floods, with recurrence rates of 1.5 times per year (Syvitski andMilliman, 2007), and modern rivers generally experience floodconditions for b10 days per year, which is less than 3% of the timeon an annual basis. Holbrook and Wanas (2014) used empiricallyestablished scaling relationships (e.g., Wright and Parker, 2004;Van Rijn, 1984) to evaluate instantaneous bedload and suspendedload discharge. These instantaneous values were then integratedover longer geological time periods, based on chronostratigraphicanalysis and sediment flux estimates. Assuming that there is goodage control, such sediment discharge estimates can then be com-pared with sediment yields in the upstream drainage basin and theamount of sediment in the downstream sinks, such as might be ob-tained from isopach maps of deltas or submarine fans (e.g., Carvajaland Steel, 2012; Guillocheau et al., 2012). Guillocheau et al. (2012)present methods to reconstruct or extrapolate cross-sectional geome-tries of stratal units in areas where distal sinks are incompletely sam-pled or imaged. Many 2D seismic lines, for example, may not samplethe entire extent of a basin. The reconstructed stratal cross-sectionscan then be used to extrapolate isopach maps,allowing total sedimentvolumes to be estimated.

In their example from the Cretaceous Bariyah Formation of Egypt,Holbrook and Wanas (2014) show that estimated sediment deliveryexceeded the calculated volumes of sediment transferred through thetrunk rivers, leading to speculation that some sediment was exportedout of the basin. Their analysis also reveals that the scale of Cretaceousrivers in Egypt was considerably smaller than the modern Nile system,and helps explain the preponderance of associated reefal carbonateformations that would likely be absent had the area been drained by amuch larger river.

Holbrook and Wanas (2014) provide a detailed analysis of the un-certainties associated with each step of their analysis, including manyof the parameters discussed above, such as channel width and depth(Fig. 13). The largest uncertainties in their study are associated withtheir empirical estimates of channel width from 1D-measured sections,as they lacked good cross sections to better constrain channel width es-timates. Another area of uncertainty is the conversion from short-termestimates of water and sediment discharge to longer-term rates(Fig. 13), which is required to link paleodischarge estimates to thestratigraphic record. It is generally known that most fluvial sedimentdischarge occurs during large-magnitude, and rare flood events, but es-timating the frequency of events is difficult in a deep-time system. Dis-charge obviously varies as function of climate regime (e.g., monsoonalversus arid, Davidson and North, 2009) but also varies with decadal cli-mate events, such as ENSO or El Nino cycles, as well as longer-termglacial Milankovitch events. Analysis of floodplain paleosols and theuse of analog regional hydraulic geometry curves, or integration withdetailed paleo-climate models may aid in making more accurateassumptions about intermediate- and long-term climate controls onsediment delivery.

6.7. Drainage area and reservoir quality from provenance analysis

Estimations of drainage basin area, lithology, locations of formerdrainage divides, and tectonic setting can be made from provenanceanalysis, a field of research that has typically focused on sandstonesrather than mudstones (Dickinson and Suczek, 1979; Mack, 1984;Dickinson, 1985). Provenance analysis bymeans of sandstone petrologyalso has important implications for estimations of reservoir lithologyand quality in hydrocarbon prospects. Volcanic arc terranes, for exam-ple, typically result in clay-rich volcanic litharenites, with an abundanceof labile framework grains, such as volcanic rock fragments, and unsta-ble minerals, such as plagioclase, mica, pyroxenes, and amphiboles(e.g., Marsaglia and Ingersoll, 1992; Critelli and Ingersoll, 1995). Thesecan undergo extensive diagenesis during burial and compaction,resulting in widespread occlusion of initial porosity forming low-quality-reservoir sandstones. In contrast, erosion of shield terranes,

such as stable continental interiors, typically results in multi-cycle,quartz-rich sandstones that generally have higher reservoir quality(e.g., Dickinson, 1985; Johnsson, 1993). Sediment composition of reser-voirs is ultimately related to its provenance,which relates to source rockcomposition (e.g., Pettijohn et al., 1987). Composition also depends onweathering and uplift rate in the source area, and weathering ofsediment in intermediate sinks (e.g., Franzinelli and Potter, 1983;Johnsson, 1990a, 1990b, 1993). For example, humid tropical climatesare typically characterized by high rates of chemical weathering, suchthat carbonate rocks experience extensive dissolution. Labile mineralsin siliciclastic and igneous rocks more readily weather to produce clay,instead of sand or gravel.

Sandstone composition is traditionally determined based on theproportion of quartz, feldspar, and rock fragments (QFL) plotted on aternary diagram (Dickinson and Suczek, 1979). Other approaches useisotopic geochemical methods to determine the geochronology ofspecific grains (e.g., feldspars, micas, apatites, and zircons) includingδ18O, 87Sr/86Sr, and 207Pb/204Pb.

Analysis of age-spectra of detrital zircons has recently become one ofthe most-used provenance analysis techniques (e.g., Rainbird et al.,1992; Gehrels et al., 1995; Dickinson and Gehrels, 2003). Because theclosure temperature of Pb in zircon is very high (N N 700 °C, Lee et al.,1997), and zircon is very chemically stable in the sedimentary environ-ment, U–Pb zircon ages are difficult to change during metamorphism,burial, or weathering. Thus, chronometric analysis of the age-range ofdetrital zircons in any given sample can fingerprint the rocks contribut-ing to the sediment in a basin at a given time. The fingerprint reflects theage of formation of the rocks or the age of formation of the provenance,in the case of zircons in sandstones (e.g., Gehrels et al., 1995). Detritalzircon age-spectra can be compared to plate tectonic and general geo-logical reconstructions to locate source areas, and thus offers greatpromise for improved quantification of drainage basin area and location

Fig. 20. Paleodrainage of continental river systems developed in the Late Jurassic and Early Cretaceous (Early Aptian). Early Aptian map is based on Blum and Pecha (2014) and Smith(1994). Jurassic paleodrainage is based on Raines et al. (2013). Green outlines of paleo-land-masses largely based on reconstructions of R. Blakey and Colorado Plateau Geosystems (2015).


in deep-time analysis. For example, the long-distance transport impliedfor grain-age distributions in Permian and Jurassic eolian sandstonesdeposited near the western shoreline of Pangaea requires the presenceof a very large, possibly multi-tributary, transcontinental river systems(e.g., Dickinson and Gehrels, 2003, 2009; Dickinson et al., 2010). Theheadwaters of these river systems can potentially be defined by integra-tion of grain ages and basement-rock distribution in time-equivalentorogenic belts or uplifted rift flanks, as inferred from extant outcropbelts (Moecher and Samson, 2006). The resulting drainage basin esti-mates, however, depend upon correct interpretation of headwatersource-rock ages and the resultant access of the drainage basins todetrital zircon grains..

Recent maps of drainage-basin evolution for North America (Figs. 7,8, 19 and 20) specify the extent of paleo-drainage basins on the basis ofdistinctive detrital zircon populations in sandstones (e.g., Dickinson andGehrels, 2003, 2009; Lawton et al., 2003; Dickinson et al., 2010;Galloway et al., 2011; Benyon et al., 2014; Blum and Pecha, 2014;Lawton, 2014). Blum and Pecha (2014) showed that the rivers ofthe McMurray Formation, which were 30 to 40 m deep and whichMusial et al. (2012) estimate to exhibit aMississippi-scale discharge, in-deed were continental-scale rivers that drained most of North America.

Riggs et al. (1996) observed 515–525 Ma zircons from the TriassicChinle Formation and Dockum Group sandstones in Texas, NewMexico, and Arizona and concluded that the trunk river had its sourcein the Amarillo-Wichita uplift of north Texas and west Oklahoma(Fig. 19). These distinctively-aged zircons are rare in rocks of other pos-sible source regions of the Chinle-Dockum system. Used in a similarway, detrital zircon ages characteristic of basement provinces in south-western Laurentia are present in Cretaceous-Paleogene sandstones ofthe Difunta Group in the foreland basin of the Sierra Madre Oriental innortheastern Mexico (Lawton et al., 2009), and indicate that a longaxial river system transported sediment parallel to the Sierra Madrethrust front from the area of the southwestern United States.

Malkowski and Hampton (2014) used U–Pb ages from detrital zir-cons in combination with a recent addition to the detrital geochemistrytoolkit: Hf isotopes. Because Hf and Zr are similar geochemically, most

zircons have a significant concentration of Hf. The isotopic compositionof Hf (187Hf/186Hf) is usually expressed in ε Hf units, which is the devia-tion of a value from the isotopic composition of chondritic meteorites.Hafnium ε units can be used as a measure of the long-term evolutionof the source rock, from which the zircons were produced. Positive εvalues indicate zircons recently extracted from the mantle, whereasnegative values indicate multi-cycle zircons derived from a cratonicsource, and extracted from the mantle during an earlier cycle of upliftand exhumation. Malkowski and Hampton (2014) analyzed zirconsfrom the Mystic assemblage (late Paleozoic) in Alaska and found agesfrom ~280 Ma to the Archean. A significant portion of these zirconshad ages from 480 to 415 Ma; these Ordovician-Silurian zircons havetwo distinct populations of Hf isotopes, one in the range + 5 to +15and the other from−3 to−25 εHf units. The two age populations indi-cate that although these zircons all came from rocks formed at about thesame time, they represent at least two distinct sources, one more juve-nile, with the positive Hf values, and the other more evolved, indicatedby the negative values.

6.8. Use of detrital zircons for formation age determination andthermochronology (uplift and exhumation rates)

Sometimes, the ages of detrital zircons can also help determine thedepositional age of a sandstone. For example, Spencer et al. (2014) an-alyzed zircons from the Van Horn Formation in west Texas. Because ofits stratigraphic position and lack of fossils, most workers had long con-sidered the VanHorn to be Precambrian. In the upper portion of the VanHorn, Spencer et al. (2014) found several Cambrian zircons, one grainwith an age of 521 ± 6 Ma. Because the U–Pb system in these zirconscould not have been reset by subsequent events, we now understandthat at least the upper part of the Van Horn must be Cambrian. Thistechnique has been used to better understand the time of depositionof other unfossiliferous rocks; in such cases, care must be taken to ana-lyze a sufficient number of zircons to give reasonable confidence thatthe extremes of the distribution have been identified (Vermeesch,2004).


Because chronometric analysis using the U–Pb method in zircons isinsensitive to the subsequent thermal history of the grains, this ap-proach is useful in identifying the formation age of the material in thesource area of a sandstone. This is, however, complicated by the possi-bility of multi-cycle grains (e.g. Campbell et al., 2005). Other chronolog-ic methods, which are sensitive to resetting at temperatures as high as400 °C to as low as ~70 °C, have been used to further understand the tec-tonics of source areas. In these cases, the apparent age of the mineraldoes not indicate the age of formation but rather the last time the min-eral passed through a certain temperature range, which can in turn belinked to its burial depth and time of subsequent uplift. The mostcommonly used isotopic tools for this sort of provenance evaluationare, in order from lowest temperature sensitivity (i.e., high closure tem-perature) to highest temperature sensitivity, muscovite (40Ar/39Ar),K-feldspar (40Ar/39Ar), zircon (U + Th/He and fission track), apatite(fission track) and apatite (U + Th/He).

Copeland and Harrison (1990) determined the 40Ar/39Ar ages ofdetrital K-feldspar and muscovite from strata of ODP sites 717–719 atthe distal end of the Bengal fan. Despite analyzing what today wouldbe considered a very small number of grains from each stratum (depos-ited from ~16Ma to ~2Ma), they found some individual muscovite andK-feldspar grains whose cooling ages were indistinguishable from thedepositional age of the fan sediment. This indicated that some portionof the source area of the Bengal fanwas being exhumed and transportedto the deep sea at very rapid rates. More specifically this gives evidencethat the closure of these systems to the loss of Ar occurred several kmbelow the surface of the Himalaya, followed by uplift, exhumation,transport, and deposition, 3000 km to the south over a very short timewindow. Although these results show that some parts of the Himalayawere experiencing this extreme rate of uplift and erosion during theMiddle to Late Miocene, it is unlikely that this is representative of thewhole mountain range. If such rates of uplift and erosion were regional,this would result in exhumation and extensive exposure ofmore deeplyburied high metamorphic grade granulite rocks in the Himalaya, butsuch granulite outcrops are largely absent.

Similar studies have been undertaken on detrital material depositedcloser to the Himalaya. For example, Bernet et al. (2006) analyzed zir-cons from ~14 Ma old to modern day foreland-basin deposits of theSiwalik Group in western Nepal by both the U–Pb and fission-track dat-ing methods. They found that the difference between the depositionalage and the youngest fission-track zircon age (closure temperature~220–240 °C) was around 4 million years, suggesting parts of westernNepal were being exhumed at rates of about 2 mm/y and transportedto downstream sinks during this time.

In theHimalaya, grainswith cooling ages equal to or very close to theage of deposition necessarily indicate order-of-magnitude faster rates oferosion (up to 1 cm/y) taking place over at least brief intervals of time.Cooling ages of grains equal to or very close to the age of depositioncould alternatively represent volcanic zircons, erupted at the time ofsedimentation. However, during the Indo-Asian continent-continentcollision, there has been little volcanism in the Himalaya, and conse-quently there is no documented volcanic detritus in the sediments ofthe Siwalik Group and Bengal Fan.

In continent-ocean convergent tectonic settings, where knowledgeof erosion rates is also of great value, interpretation of such “zero-age”grains is complicated by the possibility that the cooling age was notattained during the relatively slow transition past some criticalisotherm many kilometers below the surface, but rather during theextremely rapid transition frommagmatic temperatures to surface tem-peratures during an explosive volcanic eruption. Painter et al. (2014)addressed this problem in a study of Jurassic-Cretaceous forelandbasin deposits of the North American Cordillera in Utah, Colorado, Wy-oming, and South Dakota. (U–Th)/He ages from zircon and fission-trackages from apatite were determined from some of these sandstones. Inmany cases, the same material was analyzed for the U–Pb age as well.Grains in which the crystallization-age, as determined from U–Pb

analysis, matches the cooling age, as obtained from He or fission-trackanalysis, and is also is the same age as the sediments, are very likely vol-canic, and therefore should not be used in evaluating the erosion rates ofthe source area of these deposits. Conversely, grains in which He orfission-track ages are similar to the depositional age of the sediment,but show much older U–Pb ages (i.e., crystallization age), indicateolder grains that have not been buried for long, and these can be usedto argue for significant rates of erosion in the source area and depositionas sediments a short-time later. Painter et al. (2014) use populations ofdetrital zircons and apatites not contaminated with a volcaniccomponent to suggest erosion rates in excess of 1 mm/y in the Sevierfold-and-thrust belt during the Cretaceous.

Another concern in using thermochronology to inform us aboutthe tectonics of the source area of a sandstone, is the potential forpost-depositional resetting. This is most likely with low-closure-temperature systems, such as thefission-track and (U–Th)/Hemethods.During burial, the grains of interestmay achieve temperatures sufficientto produce open-system behavior and lessening of the apparent ages,making the ages a reflection of the thermal history of the basin, ratherthan the tectonics of the source area. Painter et al. (2014) found thisnot to be a concern for the Cretaceous foreland basin deposits in thecentral Rockies, but van der Beek et al. (2006) determined that the Neo-gene foreland basin deposits of the Siwalik Group of western Nepal hadbeen buried to a depth such that the pre-Siwalik history of many of theapatite grains has been essentially erased. As such, thefission-track agesof these apatites now reflect the exhumation of these foreland basindeposits, rather than the exhumation of the material that now fillsthese basins.

6.9. Tectonics, subsidence and sediment supply

Uplift and erosion are intimately linked processes (Beaumont et al.,1992), and of course, subsidence within sedimentary basins (which ul-timately controls preservation) includes both compactional subsidence,isostatic depression caused by sediment loading, and larger-scale litho-spheric subsidence, such asmay be related to larger-scale geothermal orplate-tectonic processes. One of the simplifying assumptions inmuch ofthe early sequence stratigraphic literature (e.g., Posamentier et al.,1988; Jervey, 1988) was of constant sediment supply, but of course innature, this is never true. Rates of erosion, and consequently sedimentproduction, may vary over an order of magnitude, especially passingfrom glacial to interglacial cycles (e.g., Hinderer, 2001), but changes insediment supply were rarely invoked to explain stratigraphic changesin sedimentary systems in the early sequence stratigraphic literature.The S2S approach has revitalized a focus on sediment supply and itsshort- and long-term controls, as indicated by the papers in this volume.

7. Examples of Mesozoic S2S systems in Canada and the USA

TheMesozoic history of North America records the pulsed evolutionof the collision and subduction of the Pacific and associated oceanicplates. This resulted in a complex history of terrane accretion in theWestern Cordillera, development of the Rocky Mountains and associat-ed foreland basins, and coincided with globally high sea-levels. Thereare corresponding profound changes in the scale, organization, and pa-leogeographic development of associated drainage basins, their rivers,and linked down-dip depositional systems. These changes also reflectvariations in play types and hydrocarbon systems. This section of thepaper reviews and synthesizes the paleogeographic history of centralNorth America, with a focus on examples from the USA and Canada.We document changes in the size of rivers, and show how changes inchannel depth and paleo-discharge can be used to interpret changesin drainage basin area and position through time, and the implicationsfor the size of down-dip depositional systems and hydrocarbonpotential.


7.1. Triassic–Jurassic

Paleo-rivers in the Triassic Chinle and Dockum formations, in thesouthwestern U.S.A. that formed during the early break-up of Pangea,largely flowed to the northwest draining Laurentian and Appalachiansource areas (Dubiel, 1992; Dickinson and Gehrels, 2008; Miller et al.,2013) (Fig. 19). Sediment contributions from western North Americanand Pacific island arc systems were minor. A major drainage divide layto the southeast (Fig. 19), roughly in the position of the older Laurentiantranscontinental arch corresponding to the Paleozoic Appalachian–Ouachita orogenic belt, which was in the early stages of rifting. Catch-ment areas are correspondingly interpreted to be on the order of6 × 104 km2 for rivers draining the Mogollon highlands to the south,and 2 × 105 km2, for rivers draining eastern Laurentian source areas.

Facies analysis of corresponding incised valley-fills in associated Tri-assic strata, such as the Dockum Formation in Texas and its equivalentsin the Chinle Formation of Utah (e.g. Dubiel, 1992, 1994), suggests thatthe largest rivers were on the order of 10–15 m in depth. These riversfed their sediments into the deforming continental margin of Idahoand Nevada (Lupe and Silberling, 1985; Dickinson and Gehrels, 2008).Extensive marine Triassic facies are also documented in westernCanada (Gibson and Barclay, 1989; Edwards et al., 1994) and thickfluvio-deltaic systems produced from draining of the northernLaurentian continent are found in Alaska (Fig. 19) (Miller et al., 2013).However, detailed paleohydraulic analyses are lacking.

The Jurassic Morrison Formation is the next major Mesozoic fluvialunit exposed in the region and reflects a significantly greater contribu-tion of sediment supplied from the emerging Western Cordilleran foldand thrust belt (Fig. 20) (Hansley, 1986). Extensive exposures, and anumber of corresponding studies focused on facies architectural analy-ses, suggest that rivers of the lower Salt Wash Member reached a max-imum of about 18 m deep (Robinson and McCabe, 1997). Owen et al.(2015) quantified changes in regional thickness of channel stories asso-ciated with the east-flowing, transverse river systems, which drainedthe rising Cordillera. Channel stories (and by inference channel depths),change from a maximum of about 7.7 m in more proximal locales to2.3 m deep in downstream areas, suggesting a distributive fluvial sys-tem with an apex near Las Vegas, Nevada. The largest transversestreamswere on the order of 8mdeep, indicating relatively small drain-age systems compared to the 18-m-deep axial river systems thatdrained the Mogollon highlands to the south.

In a study of the Jurassic rivers in the Alberta basin, which were inthe early stages of foreland basin development, Raines et al. (2013)integrated new detrital zircon age dates from the Jurassic MonteithFormation, equivalent in age to the fluvial Brushy Basin Member ofthe Upper Morrison in Utah. The Monteith includes two fundamentallydifferent fluvial systems. A larger system drained both the Mogollanhighlands to the south and an Appalachian source to the southeast,and these catchments collectively fed a north-flowing river systemthat drained parallel to the axis of the foreland basin and fed shorelinesin British Columbia (Fig. 20). Unfortunately, the record of this majorriver was largely eroded by the younger Cretaceous Cadomin Formation(Kukulski et al., 2013), but it is interpreted to have been continental inscale. A secondary source of sediment was delivered from theemerging Cordillera to the west, which formed a smaller series oftransverse drainages. Sediment derived from the western sourceareas became more dominant as the foreland basin filled. The west-ern catchment areas are estimated to have been about 100,000 km2

,

delivering sediment to coarse-grained downstream distributive flu-vial systems (e.g., Weissmann et al., 2010; Hartley et al., 2010).Channels were up to 12 m deep, and show variations in the propor-tion of channel belt versus floodplain facies (i.e., net-to-gross)through time.

From the hydrocarbon perspective, correlation and modeling ofchannel belt and floodplain stratal stacking patterns, and determinationof the size of cbannels and channel belts, were based on estimated

thickness, width and measured net-to-gross ratios, and were thenused to identify prospective sandstone-prone reservoir bodies. Thesesandstones form part of the deeply-buried tight gas deposits in the Al-berta Basin, in which a higher degree of channel-belt amalgamation isrequired to provide economic wells. Inasmuch as the thickness of chan-nel fills reflects the drainage area, estimation of channel size allow esti-mates of channel belt width to bemade, which is essential in predictingreservoir connectivity (Larue and Hovadik, 2006).

7.2. Big rivers in the early Cretaceous

The Late Jurassic drainage patterns continued into the Early Creta-ceous, and are characterized by a number of amalgamated compoundvalley systems reflecting foreland basin tectonics and the complex het-erogeneity of the North American craton, inherited from its assembly(Fig. 20). The Lower Cretaceous is marked by a major unconformity as-sociated with a number of incised fluvial units, including the Cloverly,Lakota and Dakota formations in the USA, and the Mannville Group inCanada. In the Alberta basin, a series of NW-SE trending compound val-ley systems separated by major interfluves have been identified(Fig. 20). These valleys are tens of kilometers wide and up to 50 mdeep. The easternmost Assiniboia valley is linked to the southerncontinental-scale catchment network, whereas the westernmost SpiritRiver Valley and the middle Edmonton Valley include a greater contri-bution from the Western Cordillera (Smith, 1994; Benyon et al., 2014;Blum and Pecha, 2014). Like the Jurassic systems, the larger early Creta-ceous rivers flowed parallel to the foreland basin, forming an axial sys-tem that drained the eastern Appalachian Mountains of Laurentia, andthe southern Mogollan highlands, with lesser contributions from theWestern Cordillera (Fig. 20). Coarse-grained river systems of theCloverly Formation consisted of moderate-sized rivers, up to 11 mdeep, with slopes ranging from 6× 10−3 up to 6 × 10−4, and dischargesup to 1000m3/s (Zaleha, 2013). The largest rivers in the Cloverly Forma-tion drained axially, whereas the smaller rivers drained transversely,and the overall eastward increase in discharge was partly attributed toan increase in precipitation away from the rain shadow of the EarlyCretaceous Sevier mountain belt (Zaleha, 2013).

Analysis of fluvial deposits in the supergiant heavy-oil reservesof the McMurray Formation in Canada indicate axially drained,continental-scale meandering rivers on the order of 40 m deep(Fig. 17), with paleo-discharge estimated to be on the order of15 × 103m3 (Musial et al., 2012), comparable to themodernMississippiin scale. The transcontinental drainage divide lay in the southern USAand sediment was derived from the southern extension of theAppalachians (e.g., Ouachita Mountains in Oklahoma and theMarathonuplift in west Texas, Fig. 20),). Detrital zircon analysis of McMurraysandstones indicate Appalachian and Grenville basement sources ineastern North America and corroborate the continental scale of theMcMurray drainage basin (Benyon et al., 2014).

A key aspect that affects the prospectivity of Mannville heavy oil pe-troleum systems (Fig. 17) is the heterogeneity within channels (Fusticet al., 2012). In subsurface production, oil is recovered with paired sub-surface horizontal wells using the steam assisted gravity drainage(SAGD) technology. Steam injected in the upper well liquefies the oiland allows it to drain into the lower well and then pumped to the sur-face. Extensive muddy inclined heterolithic layers (IHS), associatedwith point bars (Fig. 17C) are barriers that impede steam injectionand inhibit flow of the oil, and prediction of these barriers is thus critical(Thomas et al., 1987). In areas where oil sands are mined, the IHS mud-stone layers slow the mining operation, as they do not contain oil andmust be excavated and disposed of before processing of sands can re-sume. Understanding that Mannville point bars are formed by acontinental-scale river is reflected in the large-scale of channel belts.As a consequence, individual inclined layers are laterallymore extensivethan the typical well spacing, thus allowing them to be correlated andmapped explicitly, versus modeled stochastically (e.g., Fustic et al.,


2013). Muddy inclined drapes may form as counter-point bars in up-stream reaches (Smith et al., 2009) and by tidal processes in moredownstream reaches (Thomas et al., 1987). In the more-distal reachesof the ancient river systems, within the backwater limit, tides weremore important as reflected in the abundance of bioturbation andothermarine indicators. Given the continental scale of the river systems,gradientswere certainly low, andprobably on the order of 10−4 (Musialet al., 2012). Backwater lengths would correspondingly have been onthe order of 400 km, allowing tidal effects to propagate far upstream.Regional paleogeographic maps of Smith (1994) show that the heavyoil deposits in the McMurray were deposited within a few hundred ki-lometers of the shoreline and were thus likely within the backwaterlength of the system. Blum (2015) in contrast, used 3D seismic imagesto document the migration patterns of Mannville channels and sug-gested that the high sinuosity and the high degree of lateral migrationofmeanders require the channels to have been landward of the backwa-ter reach and thus beyond the limit of tidal influence, although this is in-compatible with the observation of marine microfossils in associatedmudstone drapes.

TheMannville valleys includemajor axial systems, aswell as smallersteeper-gradient transverse tributary systems, draining the WesternCordillera, which fed the extensive shorelines of the sand-and gravel-prone Fahler Formation and associated prodeltaic and shelf mudstonesof the Spirit River Formation in Northern Alberta and British Columbia.These reach up to 300 m thick. The Edmonton valley system also fedshorelines of the Bluesky Formation in northern Alberta and BritishColumbia.

The Fahler shoreface and deltaic sandstones form individualparasequences, tens of meters in thickness, and cover areas on theorder of 3 × 104 km2 (Leckie, 1986; Hayes et al., 1994). The areal scaleof the associated sandydelta systems (Fig. 20) is broadly similar tomod-ernMississippi or Niger, but thickness is quite a bit less, and is consistentwith the smaller catchments of the Western Cordillera. Isopach sand-stone volume of an individual mapped deltaic system of the Fahler(i.e., a parasequence) is estimated to be on the order of 3 × 102 km3.Total sedimentary volume, including mudstone and sandstone, is likelyan order ofmagnitude greater (e.g., 103 km3). Paleogeographicmappingby Smith (1994) demonstrated that Upper Mannville shorelinecomplexes of the Bluesky Formation extended along strike for over1000 km and along dip for 600 km, suggesting widespread shorelinedeposits consistent with sediment supply associated with the largerriver systems of the Assiniboia valley.

There is good evidence of numerous high-frequency rises andfalls of sea level during the Early Cretaceous forming individual del-taic parasequences (e.g., Leckie, 1986; Hayes et al., 1994). TheMannville Group contains a number of important transgressive epi-sodes (Smith, 1994; Hayes et al., 1994). At times shorelines lay incentral Alberta, and at other times shorelines prograded several hun-dred kilometers seaward into British Columbia and beyond. Unfortu-nately, the distal limits of deep-water systems fed by the largestMannville rivers are largely eroded as a result of later forelandbasin tectonics and Quaternary glacial erosion, consequently no re-cord exists of linked shelf clinoforms or deeper-water submarinefans. Much of the distal sediment delivered to this Seaway is eithernow contained within the Rocky Mountain fold and thrust belt, oreroded away. However, comparison with the volumes of sedimenton modern continental-scale submarine fans, such as the Mississippi(Walker, 1992) may be used to suggest that Mannville rivers deliv-ered Mississippi-scale volumes (3 × 105 km3) to the Boreal Seaway,and would indicate that the bulk of sediment delivered to theocean by the Mannville Rivers was transferred out of these systemsand is thus no longer preserved in the rocks of western NorthAmerica.

Therewere certainlymuds delivered to the shelf byMannville rivers,but these did not serve as the major source of heavy oil for theMannville, although they may source hydrocarbons in the Bluesky and

Fahler shoreline sandstones farther west. The source rocks that chargedthe Mannville heavy oil deposits are thought to be Jurassic to Paleozoicin age (Creaney et al., 1994). These source rocks matured in the deeperparts of the foreland basin, producing oil that migrated up-dip,across the erosional valley floor, and into the Manville. In terms ofpetroleum systems, the oil-sands source rock is disassociated(i.e., derived from older formations), although locally, the MannvilleGroup does contain self-sourced gas, some oil, and extensive coal, es-pecially in themore deeply-buried western parts of the Alberta basin(Hayes et al., 1994).

In terms of the heavy oil resource, evaporation and degradation ofthe oil at the surface and in the shallow subsurface dispersed the lighterhydrocarbon compounds leaving heavy oil in the reservoirs (Creaneyet al., 1994). Because the deposits are shallow, they are easily accessible,but the high viscosity results in difficulty in flowing the oil requiring ex-tensive reservoir engineering, such as the use of horizontal wells andsteam-assisted gravity drainage (SAGD). The scale of the river systemsand the associated heterogeneity is a major factor in determining theeconomic viability of this vast resource.

7.3. Late-Early Cretaceous, end of the big rivers

The top of the Mannville, at around 100 Ma, is marked by a regionaltransgression, associated with high global sea level and flexural depres-sion of the North American Plate, enhancing the developing forelandbasin. The transgression is marked by the Joli Fou Shale in Canada andits equivalents in the USA (Reinson et al., 1994). This transgressiveshale forms a major hydraulic barrier to oil and gas migration suchthat no oil and gas from older source rocks migrates across this shaleboundary (Creaney et al., 1994). Despite this transgression, riverssystems in the overlying formations also flowed northwest, deliveringsediment into the Boreal Seaway. Incised valleys in the VikingFormation and Paddy Member of the Peace River Formation in Canadawere associated with a Late Albian lowstand, and are equivalent in ageto the Muddy Sandstone in Colorado and Wyoming. U–Pb dating ofzircons from the Paddy Member (Buechmann, 2013) shows similarprovenance to the Mannville, with Laurentian and Cordilleran sourceareas. The Paddy valley was large, tens of kilometers wide, andhundreds of kilometers long. Lateral accretion bedding and associatedtidally influenced point bars are on the order of b10m thick. Abandonedchannel mud plugs in the Paddy are about 7 m thick (Leckie and Singh,1991), indicating rivers were significantly smaller than the Mannvilletrunk streams (b10 m deep) and suggesting a less well-integrated,probably more segmented drainage network. Outcrops of the Vikingare relatively rare, and a comprehensive analysis of river scales hasnot been attempted.

The western Canada foreland basin was formed above a heteroge-neous lithosphere characterized by a number of subtle arches, such asthe Peace River and Sweetgrass structures, and at times these causedsegmentation of the basin and associated drainages, especially atlowstand when the basin was characterized by fluvial deposition.Motion on these structural lineaments may reflect intra-plate stressesor activation of a peripheral bulge (e.g., Stelck, 1975; O'Connell, 1994;Wright et al., 1994; Plint and Wadsworth, 2006).

The sandstones of the Viking and Muddy formations constitute im-portant oil and gas reservoirs in Canada and the USA, including fieldswithin incised valleys and shallow marine facies (Dolson et al., 1991;Reinson et al., 1994). The depth of incision of valley fills is typically asso-ciated with shoreface deposits, exposed during falls of sea-level(e.g., Wescott, 1993). The exposed wedge of highstand sediment typi-cally has a steep-fronted shoreface, forming a nickpoint that allows inci-sion of about 20 m. Variation in the width of valley-fills, and theirassociated reservoirs, is controlled by the duration of sea-levellowstands (Wescott, 1993). Longer-lived lowstands result in widerand more complex compound valley fills, as appears to be the case ofthe older Mannville and Paddy valleys, which were tens of kilometers

Fig. 21. Cenomanian paleogeography of North America. Small, high relief drainage basinsare linked to an activemountain chain. Drainage areas are estimated from paleogeograph-ic and paleotectonic reconstructions. Seaway was closed to the south. Cretaceous Seawaybased on Bhattacharya and MacEachern (2009). Woodbine systems is based on Adamsand Carr (2010). Green outlines of paleo-land-masses largely based on reconstructionsof R. Blakey and Colorado Plateau Geosystems (2015).


in width. In contrast, smaller valleys in the Viking were typically only afew kilometers in width and may indicate shorter lowstands (Reinsonet al., 1994).

7.4. Late Cretaceous (Cenomanian), rivers and deltas of the DunveganFormation and equivalents

In general, the Cenomanian marks a time of segmentation of drain-ages in the Seaway and a profound transition from northwest- tosoutheast-flowing rivers (Fig. 21). Tectonic salients and promontoriesin the Sevier orogenic front, aswell as the tectonic lineaments discussedabove, exerted a control on localization of discrete delta systems andtheir fluvial feeder systems, such as the Dunvegan Formation inCanada and the Frontier Formation in Wyoming (e.g., Lawton et al.,1994). The transcontinental arch was still a positive feature at thistime, and the Boreal (northern) Seaway is interpreted to have beenclosed to the south (Williams and Stelck, 1975). The confined southernarm of the Seawaymay have led to anoxic substrate conditions forminghydrocarbon-rich source rocks, including the Fish-Scales unit in Canadaand the Mowry Shale in the USA.

The Woodbine Formation in Texas represents a modest-sized deltasystem on the southern side of the drainage divide that drained the Ap-palachians and prograded to the southwest into the paleo- Gulf ofMexico Seaway (Fig. 21). Rivers in the Woodbine were small, less thana few meters deep, reflecting a localized drainage area (Oliver, 1970;Ambrose et al., 2009; Adams and Carr, 2010), but rich source rocksensure that the Woodbine is the host reservoir for the giant East Texasoil fields.

7.4.1. The Dunvegan deltaThe Cenomanian Dunvegan Formation, in Canada marks the first

appearance of major Cretaceous rivers that flowed southeast (Figs. 21and 22), representing a 180° paleogeographic reversal from the preced-ing Viking and Paddy shorelines. Conglomeratic alluvial facies of theDunvegan are exposed in the Northwest Territories of Canada (Stott,1982), and record accretion of continental terranes to northwesternNorth America (Johnston, 2008).

Extensive outcrop and subsurface well data through the Dunvegan,allow mapping of individual valleys and their associated down-dipsandy delta lobes and prodeltas (e.g., Bhattacharya, 1993, 1994) en-abling a comprehensive S2S analysis (Figs. 22–25). Extensive subsurfacemapping shows that the Dunvegan valley systems form an extensivetributary network that coalesce into trunk valleys (Plint, 2000, 2002;Plint and Wadsworth, 2003, 2006) (Fig. 22). Mapped individual valleysegments, including the trunk valley, are typically less than a fewkilometers in width, and the depth of incision was generally less thanabout 35m. Age-equivalent deltaic sandstones and prodeltamudstonescan be mapped downlapping onto a distal peripheral bulge (Plint et al.,2009; Fig. 25), in a mostly closed depositional sink that allowsestimation of bulk sediment volumes.

Allostratigraphic analysis shows that the entire Dunvegan Formationwas deposited in about 2 million years. The ten mapped allomembersrepresent deposition in no more than 200,000 years, and probably asfew as 100,000 years. Individual parasequences likely represent deposi-tion in much less than 50,000 years, assuming depositional rates wereon the order of 1–10 cm/y. These chronometric constraints allowlong-term sediment budgets to be estimated.

Plate tectonic considerations (Plint and Wadsworth, 2006) and re-cent detrital zircon dating (Buechmann, 2013) show the bedrock sourcearea was about 150,000 km2 (Fig. 22). Outcrops and cores samples ofvalley fills were used to document key paleohydraulic parameters(Figs. 24 and 26). Outcrops of tributary channels show well-developedlateral accretion, indicating single-thread meandering rivers, ~16 mdeep and ~150 m wide (Fig. 24). Trunk streams must have been corre-spondingly larger, and perhaps closer to 20 m deep, although there areno outcrop exposures of the trunk-valley rivers, so width of trunk riversis estimated from empirical equations.

Trunk valley fills are filled primarily with decimeter-thick sets ofdune-scale cross-bedded fine- to medium-grained (125–350 μm) sand-stone (Fig. 26). Using the bedform phase diagram, as described inSection 6.4,flowvelocities on the order of b1.5m/s have been estimated(Bhattacharya and MacEachern, 2009; Fig. 15).

Plint et al. (2009) used correlations and modern analogs to estimateregional paleo-slopes for the Dunvegan rivers of 3–4 × 10−4, suggestinga moderate- to low-gradient system. Recalculations of slopes, based ongrain-size analysis and water depth estimates (using the Manning andBrownlie equations as described in Section 6.4) indicate a significantlylower slope of 3–5 × 10−5. Using the higher slope and assuming channeldepths of up to 20 m, backwater lengths of about 50–65 km areestimated. Using the lower slope estimate yields a backwater length of400–600 km.

Regional considerations, including plate-tectonic position from pa-leomagnetics aswell as analysis of floodplain paleosols, indicate deposi-tion at about 70° latitude, within a temperate-humid climate regime(Irving et al., 1993; McCarthy and Plint, 1999; McCarthy, 2002). Theelongate, axial drainage area was about 1.25 × 105 km2 (Fig. 22A).

7.4.2. Dunvegan paleodischarge estimatesBased on flow velocities determined from the sedimentological

observations (grain size, channel thickness, channel width, and sedi-mentary structures), and using the bedform phase diagram approach(Fig. 15), bankfull paleodischarge of Dunvegan rivers was estimated tobe 4–6 × 103 m3/s (see Table 1). Paleodischarge estimates of3.5 × 103 m3/s–1 × 104 m3/s, were based on regional hydraulic geome-try curves (Davidson and North, 2009) assuming a drainage basin of

Fig. 22.Dunvegan paleogeography, A. tectonic regime and drainage basin (125,000 km2), B. Tributive valley systems feed trunk rivers andmajor delta lobes, based on over 3000well logs,cores and outcrops in fold and thrust belt (after Plint and Wadsworth, 2003, 2006). Cross section A–A′ shown in Fig. 23.


about 1.25 × 105 km2, and are consistent with sedimentological obser-vations. A discharge estimate of 6 × 03 m3/s was calculated using theManning Equation (with n = 0.035). The corresponding velocity esti-mate of 2.4 m/s, was based on a 20 m deep, 200 m wide meanderingchannel.with a calculated hydraulic radius of about 11 m, flowing overa slope of 3 × 10−4. This velocity is well within the range predictedusing the other techniques, such as the bed form phase diagram. Usingthe lower slope (3–5 × 10−5) yields a velocity of 0.8 m/s and a Qw of1600–2300 m3/s, which is a factor of 3 lower. In general, the dischargeestimated using sedimentologic criteria, such as channel depth, grain

Fig. 23. Depositional dip cross section through the Dunvegan Formation, Alberta, Canada.Bhattacharya and Walker (1991).

size, and slope, as outlined in Section 6 range from 1.6–6 × 103 m3/s,or over a factor of 3.75. The largest deviation appears to be associatedwith use of the regional hydraulic geometry curves (factor of 6.25difference compared to lowest estimate using sedimentologicalcriteria), andwe suggest that this technique has the largest uncertainty.

Bhattacharya and MacEachern (2009) used paleodischarge esti-mates of Dunvegan trunk rivers, and compared these to empiricalrelationships based on modern rivers to predict their propensity forgenerating hyperpycnal flows (Mulder and Syvitski, 1995). Mulderand Syvitski (1995) showed that small rivers, with an average discharge

Sequence boundary in Allomember E dives under attached lowstand delta (E1). From

Fig. 24. Large-scale lateral accretion sets in a point bar within a Dunvegan valley fill. Photo of cliff section (above) and interpreted bedding diagram (below) Lowest panel shows recon-structed channel (modified after Plint and Wadsworth, 2003).


of b6000m3/s, routinely produce hyperpycnal flows during subsequenthigher-discharge floods because the suspended sediment concentra-tions typically exceed the capacity of the flow. In contrast, they showthat rivers with average discharge, N1 × 104 m3/s, never producehyperpycnal flows, even during exceptional floods, because thesuspended sediment concentration is unable to reach the requiredthreshold. The estimates of paleo-discharge for the Dunvegan, likely re-flect flood conditions, rather than average flow conditions (see discus-sion in Bhattacharya and MacEachern, 2009), but in either case do notexceed the 6000 m3/s threshold of Mulder and Syvitski (1995). Wheth-er or not a given paleo-river actually generated hyperpycnal flows can

Fig. 25. A. Isolith map of Parasequence E1 in Allomember E, Dunvegan For

be evaluated by examining associated delta front and prodelta facies,which may record sediment delivery mechanisms, especially in settingwhere sedimentation rates are high and levels of bioturbation low(e.g., Bhattacharya and MacEachern, 2009; Li et al, 2015; Keuhl et al.,2015; Romans et al., 2015). In the case of the Dunvegan, examinationof prodeltaic facies in core shows low levels of bioturbation, rare waveripples and some hummocky cross stratification (HCS), and abundantinverse and normal graded beds with current ripples and Bouma se-quences. The inverse graded bed and Bouma sequences are diagnosticof hyperpycnal flows and turbidity currents respectively (Mulderet al., 2003). The low bioturbation indicates high sedimentation rates

mation. B. Isopach map of Allomember E. From Bhattacharya (1993).

Fig. 26.Measured sections through tributaries in Dunvegan valley fills. Data allow measurement of grain size, bedforms, and story thickness. Fining-upward channel stories indicated byarrows. Modified after Plint and Wadsworth (2003).


and suggests a river-dominated prodelta. The wave ripples and HCS in-dicates periodic storms that aided in along-shelf and across-shelf shelftransport (Plint et al., 2009)

7.4.3. Dunvegan sediment budgetsIsolith mapping of sandy delta deposits within various allomembers

(Fig. 25) shows that individual lobes (i.e., parasequences) cover an arearanging from 2 to 5 × 103 km2, with sandstone thickness up to 15 m(Bhattacharya, 1993). Individual delta lobes thus contain sand volumeson the order of 10 km3. Isopachmaps of Dunvegan allomembers, whichinclude sandstone and shales, indicate sediment bodies that are about200 km in length, from their updip to down-dip pinchout limit, and ex-tend along depositional strike for at least 300 km, representing an areaof at least 6 × 104 km2 (Fig. 25). Allomember E is about 50 m thick, andthe totalmapped sedimentary volume (including porosity) is estimatedto be about 3000 km3. The landward to seaward extent is wellconstrained by the onlap and downlap limits of Allomembers, thus de-fining a closed system, although the southwestern limit has not beenconstrained and thus it is possible that along-shelf export of sedimentmay have occurred.

Time stratigraphic analysis suggests that Dunvegan valleys mayhave persisted for 10,000–50,000 years. Fulcrum analysis (i.e.,Holbrook and Wanas, 2014) of the Dunvgen valley fills integrating in-stantaneous discharge estimates over these longer time periods, sug-gests that total sediment delivery was between 200 and 800 km3 ofsediment. This would correspond to a single parasequence within anallomember. The rock rock volume of Parasequence 1 in AllomemberE, which represents the lowstand delta deposits fed by the valley, is

about 750 km3 (i.e., about 1/4 of the total sediment volume in theAllomember), and matches remarkably well with the volume of sedi-ment estimated to be delivered out of the trunk valley.

7.4.4. Dunvegan petroleum perspectiveFrom the petroleum perspective, the Dunvegan Formation contains

numerous small conventional andmature oil and gasfields, primarily as-sociated with incised valleys and lowstand delta lobes (Bhattacharya,1994). Because the Dunvegan progrades southeast but the basin isstructurally tilted to the southwest, traps typically require a local lateralstratigraphic pinchout to the northeast. Such trapping configurationsoccur along the northern margins of incised meandering valleys ornorthern margins of delta lobes, resulting in localized traps and smallfields. More recent exploration in the Dunvegan Formation is focusedon unconventional resources in the down-dip heterolithic prodelta fa-cies. The nature and productivity of these thin-bedded reservoirs is con-trolled by the geomechanical properties, which in turn were created bythe depositional processes. For example, knowing the proportion ofpreserved hyperpycnal deposits, and turbidites, versus reworkedstorm beds or more slowly deposited and typically more bioturbatedsuspension deposit may control the sorting, thickness, grain size andand textural properties of beds, which are in turn important forpredicting the productivity of horizontal wells and associated fracturejobs in these thin-bedded reservoirs. The approach presented abovemay help to predict the propensity of a particular deltaic delivery pro-cesses (e.g., hypercnal versus hypopycnal flows) and, when integratedwith knowledge of shelf processes (e.g., storms, oceanic currents) and

Fig. 27. North American paleogeography during the Turonian. A series of separate deltasystems drained portions of the Western Cordillera. Drainage Areas are 50,000 km2. K-Kaiparowits. NM – New Mexico. Based on Gardner, 1995 and Bhattacharya and Tye,2004. Green outlines of paleo-land-masses largely based on reconstructions of R. Blakeyand Colorado Plateau Geosystems (2015).


sedimentation rates may help determine the nature of prodelta faciesand along-strike variability (Li et al., 2015).

7.5. Turonian: a time of many deltas

The Turonian age was marked by an overall expansion of theCretaceous Seaway, which finally connected the newly formed Gulf ofMexico andAtlantic Oceanwith the Boreal Seaway to the north. A seriesof separate S2S deltaic-shoreline systems persisted along the westernmargin of the Seaway (Fig. 27).

One of these includes the Cardium Formation in Alberta, whichcomprises a series of parasequences deposited in broadly elongatewave-dominated shorefaces fed by numerous east-flowing rivers (e.g.Krause et al., 1994 and references therein; Plint et al., 1988). Becauseof the paleogeographic reorientation, the wave-dominated shorelinesands of the Cardium Formation pinch out to the northeast, in the oppo-site direction to the tectonic tilt of the foreland basin. This produces anear-perfect structural/stratigraphic trap, resulting in giant oil and gasaccumulations. This is in distinct contrast to the Dunvegan, where thepaleogeographic dip parallels tectonic strike resulting in much smalleraccumulations, despite probably being fed by a larger river.

The Cardium parasequences were likely fed by numerous smallrivers, which flowed transverse to the foreland axis. Localizedconglomerates cap several of the parasequences, and have beeninterpreted to mark entry points of gravelly-bedload rivers (Walkerand Eyles, 1988; Arnott, 1991, 1992; Hart and Plint, 2003). The abun-dance of conglomerate suggests that rivers were steep-gradient(e.g., Slope ~10−3), and the thickness of fluvial conglomerates suggestchannels were b6 m deep (Hart and Plint, 2003). Wave reworkingformed shorefaces that were amalgamated along strike, resulting in arelatively homogenous reservoir body, locally capped by the conglom-erates (Bergman and Walker, 1987). The conglomerates also mark

regional transgressive marine erosion surfaces, which supports the in-terpretation that rivers were shallow, as deeper channels would havebeen more likely to survive the transgression (Bhattacharya, 2011).

In addition to the sandstones, the marine conglomerates in theCardium form prolific hydrocarbon reservoirs (Krause et al., 1987,1994). Exploration efforts in the Cardium are now focused on thin-bedded facies on the seaward margin of the shoreface sandstone body.Along-strike variability in these facies likely depend on the positionand lateral spacing of fluvial entry points, which should be marked bylower bioturbation intensities and an increase in beds formed by riverplumes.

The Frontier Formation in Wyoming (Gani and Bhattacharya, 2007;Lee et al., 2007; Vakarelov and Bhattacharya, 2009) similarly formed aseries of extensive wave-dominated and locally conglomeraticshorefaces with local deltaic input (Fig. 27). Like the Cardium, steep-gradient, small rivers are indicated by the abundant transgressivelyreworked conglomerates that cap sandstone bodies. In addition to thesandstones, the conglomerates create locally prolific reservoirs

The Ferron Sandstone Member in Utah (detailed below) and theGallup Sandstone Formation in NewMexico are more deltaic in nature.River deposits are preserved and both systems show an axial compo-nent, despite their catchments being contained entirely within theCordillera (Fig. 27). Sediment sinks derived from eastern NorthAmerica were confined to the eastern margin of the Seaway, as indicat-ed by the appearance of Appalachian-derived zircons in the Dakotasandstone in Iowa and Nebraska at the eastern Seaway margin asearly as late Albian time (Finzel, 2014).

The Cardium, Frontier, and Gallup all contain important oil, gas, andin some cases, coal deposits, and have all received extensive study(e.g., Plint et al., 1986; Krause et al., 1994; papers in Van Wagoner andBertram, 1995). Source rocks for these formations are derived fromunderlying and overlying condensed section shales. These formed atmaximum highstands, and probably were enhanced in organic matterby restricted circulation in the Seaway and development of anoxicsubstrates (e.g., Schlanger and Jenkyns, 1976; Jenkyns, 1980; Creaneyet al., 1994).

7.6. The Ferron sandstone, a case study

One of the most extensively studied Turonian units from a S2S per-spective is the Ferron Sandstone Member. The Ferron Sandstone, amember of the Mancos Shale Formation, comprises a series of threebroadly deltaic clastic wedges fed by rivers draining the emergentpaleo-Rocky Mountains during the mid- to late-Turonian (Garrisonand van den Bergh, 2004). The Notom, Last Chance, and Vernal deltacomplexes have been recognized (Cotter, 1975), and the Last Chanceand Notom have been extensively studied (Fig. 27; see papers inChidsey et al., 2004; as well as papers by Fielding, 2010, 2011, 2015; Liet al., 2010; Zhu et al., 2012; Li and Bhattacharya, 2013; Ullah et al.,2015). Paleosol and palynological analyses indicate that the climatewas subtropical and probably ever wet (Akyuz et al., 2015).Bhattacharya and Tye (2004) estimated the area of Ferron drainagebasins as approximately 5.0 × 104 km2 (Fig. 27).

Detailed sequence-stratigraphic studies suggest incised valleysoccur with the Notom and Last Chance systems (see Li et al., 2010 andGarrison and van den Bergh, 2004; Fig. 15), and detailed facies-architecture studies have documented the style and scale of channelsand channel belts (e.g., Corbeanu et al., 2004; Li et al., 2010;Bhattacharyya et al., 2015; Ullah et al., 2015; Wu et al., 2015;), allowingtrunk rivers to be pinpointed (Figs. 10 and 14).

The depth of the channels in the Last Chance delta complex wereabout 11 m at their deepest (Garrison and van den Bergh, 2006), andthese liewithin valleys exhibiting up to 32mof erosional relief. Channelwidths, measured from outcrop data, range from 100 to 174 m.Channel-belt sandstones are dominated by medium-grained, cross-bedded sandstones, which are locally conglomeratic. Extraformational


conglomerate phases are typically pea-grade gravel, and indicate thatFerron rivers drained the Sevier hinterland. Valleys within the Notomwedge also show relief of up to about 30 m, but they are up to severaltens of kilometers wide and compound in nature suggesting a long-lived and complex history associated with prolonged falls andlowstands of sea-level (Li et al., 2010; Li and Bhattacharya, 2013).Slope estimates in theNotomdeltaic complex associatedwith these val-leys are on the order of 10−3, and were made based on onlap-distanceand height of aggrading coastal prisms, as well as the elevation dropof valley floors along a depositional dip-oriented regional cross section(Fig. 15B). These slopes suggest a relatively steep fluvial gradient.

Plan-view exposures of channels and channel belts (Fig. 11C) indi-cate low-sinuosity (b1.44), single-thread, meandering sand and gravelbed rivers, containing bars built mostly with medium-grained cross-bedded sandstones (Fig. 12). Paleo-thalwegs of channels within valleysin the Notom system appear to be about 5–7 m deep (Li et al., 2010;Ullah et al., 2015.), with average channels 2–3 m deep. Plotting these

Fig. 28. Paleogeographic reconstruction of a valley-fed delta, Paraseq

on the bedform phase diagram indicates flow velocities of between0.75–1.3 m/s (Li et al., 2010) (Fig. 16). Calculations using the ManningEquation yield velocities of up to 2.3 m/s. Assuming widths of 100–174 m, Li et al. (2010) estimated bankfull paleodischarge of the trunkrivers to be 1.3–3.0 × 103 m3/s. Davidson and North (2009) used hy-draulic geometry curves to estimate discharges ranging from 66 to2833 m3/s. Bhattacharya and MacEachern (2009) employed these esti-mates to predict that even the largest Ferron rivers were routinely capa-ble of generating hyperpycnal flows during high magnitude floodevents, suggesting high rates of delivery of sediment to the shelf. Thishypothesis is also consistent with the observation of inverse-gradedhyperpycnites as a common component of prodeltaic facies within theunderlying Tununk Shale, fed by Ferron rivers (Bhattacharya andMacEachern, 2009; Li et al., 2015).

We used the fulcrum technique of Holbrook and Wanas (2014) toestimate both short- and long-term sediment discharge rates and con-sequent sediment volumes delivered by Ferron paleorivers (Table 1).

uence 6 of the Ferron Notom wedge (from Ahmed et al., 2014).


Slopes of 0.7–0.8 × 10−3, calculated based on grain-size analysis, aresimilar to slope estimates using stratigraphic considerations above.The larger rivers in the youngest sequence of the Ferron Notomwedge, show a paleodischarge of 1.3 × 103 m3/s bedload discharge of0.7 m3/s and a suspended-load discharge of 10.5 m3/s. We assumethat Ferron rivers flowed at bankfull rates for about 10–14 days peryear, comparable to modern sub-tropical rivers. This yields an annualmean bedload estimate of 9 × 105 m3/y. Owing to the coarse grade ofthe paleo-river bedload, we used the equations of Wright and Parker(2004) rather than Van Rijn (1984) to calculate suspended load.Assuming deposition occurred across one 20,000-year Milankovitchcycle, results in a total bedload sediment volume of about 20 km3 anda total sediment volume of about 300 km3. Correlation and mappingof sandstones delta lobes within Ferron parasequences show that theywere approximately 500–1000 km2 in area and about 10–20 m thick(Figs. 15 and 28), representing a sandstone volume of 5–20 km3,which matches remarkably well with the bedload estimate above.Unfortunately, the limits of prodelta and shelf mudstone facies withinindividual Ferron parasequences have not been mapped, but if theseare assumed to be roughly an order of magnitude more than thebedload, total sediment volumes would be 50–200 km3, consistentwith the estimates made above using the fulcrum method (Holbrookand Wanas, 2014).

The Ferron has long been regarded as an analog for paralic petro-leum reservoirs worldwide (Bhattacharya and Tye, 2004; Deveugleet al., 2014), but there has been debate as to the scale of Ferron riversand suitablemodern analogs. The Ferron is also important in containingextensive coals deposits and some gas (see papers in Chidsey et al.,2004). The coal deposits are largely associated with times of positiveaccommodation, and likely formed in wetland environments behindlinked shoreline deposits that that show an aggradational toretrogradational component to their shoreline trajectory.

The Ferron and its age-equivalents along the Cretaceous Interior Sea-way are overlain by deposits of a major transgression (Niobrara trans-gression of Kauffman, 1977). The transgressive strata include the BlueGate Member of the Mancos Shale Formation in Utah, the NiobraraShale in Wyoming and Montana, and the First White Speckled Shale inWestern Canada, and correlative strata. Within these shales, a numberof stacked thin shoreline units are documented along the western mar-gin of the Seaway (e.g., the Emery Sandstone in Utah and the MedicineHat Sandstone in Canada), but nomajor fluvial deposits of this age havebeen described. However, the cyclic alternation of progradational clas-tic, fluvial-deltaic wedges and transgressive organic-rich shales formnumerous reservoir, source and seal triplets that make up the extensiveCretaceous petroleum systems along the margins of the CretaceousInterior Seaway.

Presently, greater attention is being paid to these shalier formationsas they are the target for unconventional oil and gas reserves. Key toproduction is the ability to fracture the shaly formations. The richestsource rocks commonly are associated with marine highstand con-densed sections, but these contain abundant bentonites and high organ-ic content, which results in elastic rocks that do not easily fracture.Downlapping river-fed prodeltaic or along-strike mud-belt facies,whichprograde over these condensed sections,while leaner in organics,may contain higher proportions of silt and quartz, improving theirgeomechanical properties for induced fracturing. Superposition ofdownlapping river-derived prodeltaic deposits (which also have betterprimary porosity) with pelagic and hemipelagic source intervals mayyield more prospective unconventional reservoir-source pairs.

7.7. Santonian to Paleocene, the mega-bajada

The next major phase of recorded fluvial deposition (and last fordiscussion) began in the Santonian and continued throughout the re-mainder of the Cretaceous Period with two major cycles, an earlierSantonian to Campanian cycle (~84 Ma) and a younger Maastrichtian

to Paleocene phase (~70–60 Ma) interrupted by transgression of theBearpaw-Clagget shale (Kauffman, 1977).

Thick Campanian fluvial successions include the uppermost StraightCliffs andWahweap Formations in southern Utah (Eaton, 1991; Lawtonet al., 2014a; Lawton et al., 2014b), the Mesaverde Group, including theBlackhawk-Castlegate Formations in central Utah (Hampson et al.,2013; Hajek and Heller, 2012), the Ericson Formation in Wyoming(Martinsen et al., 1999), the Williams Fork Formation in Colorado(Pranter et al., 2009), and the Belly River/Judith River in Canada.Maastrichtian fluvial successions include the Lance Formation inWyoming (Carvajal and Steel, 2009) and the Horseshoe CanyonFormation in Canada (Dawson et al., 1994). The Campanian unitsformed a relatively contiguous band forming a giant bajada-like alluvialfringe, a few hundred kilometers wide, which likely extended for sever-al thousand kilometers alongmost of thewesternmargin of the Seaway(Fig. 8A). Paleo-river systems in southern Utah comprise transverse dis-tributive fluvial fans, eroded from the thrust-belt, which connectedwithnorth flowing axial rivers that drained the rejuvenated Mogollon high-lands in Arizona (Lawton et al., 2003). Detrital zircon studies indicatethat these north-flowing axial systems drained basement rocks thatlay south of the Cordilleran foreland basin (Lawton et al., 2014a,2014b; Szwarc et al., 2015). The sand-rich distributive fluvial fans ofthe Straight Cliffs andWahweap Formations comprised upstream fluvi-al systems that were sourced from weakly-lithified Mesozoic eolianunits in the frontal parts of the Sevier orogenic belt, with erosion en-hanced by a monsoonal climate (Lawton et al., 2014a, b). The extensiveuplift was likely caused by the overall transition into thick-skinnedtectonics of the Laramide orogeny, forming an extensive western pla-teau which culminated in the Eocene-age “Nevadaplano” of DeCelles(2004). The transverse component became increasing important asdeformation progressed (Lawton et al., 2014b). Eventually, Laramidedeformation caused the Cretaceous Seaway to retreat, and the forelandbasin was segmented into a series of largely non-marine CenozoicLaramide sub-basins.

Campanian rivers in Central Utah, Wyoming and Canada flowedbroadly east to southeast, reflecting dominantly transverse drainagewith little axial transport. Hampson et al. (2012, 2013) and Floodand Hampson (2015) compiled data on the facies architecture andpaleodischarge of river deposits in the Blackhawk Formation in Utahalong an approximately 100 km stretch of near-continuous exposuresand also attempted a mass-balance S2S analysis (Hampson et al.,2014). Single-thread meandering rivers exhibit a maximum depth of9.8 m. Multi-thread, braided rivers show depths ranging from 2.4 to14.2 m. Paleoslope estimates of 0.7 × 10−3, suggesting steep gradients.Estimates of peak water discharge range from 0.42–2.2 103 m3/s formeandering streams and 9.0–11 × 103 m3/s for the braided channels.The largest values are similar to themodernMississippi, but almost cer-tainly drained a smaller area. Numerous rivers, spaced several tens up toabout 70 km apart, fed sediment to the shoreline and were reworkedinto an extensive, and relatively smooth, wave-dominated coastline,similar to the Cardium Formation discussed in Section 7.5. TheBlackhawk alluvial plain thus consisted of a series of separate riversand floodplains, probably drained from separate segments of theWest-ern Cordillera, but coalescing to form a single amalgamated coastalplain. Collectively, these rivers would have supplied massive amountsof sediment to the Cretaceous Seaway. The Belly RiverWedge in the Al-berta basin covers about 500,000 km2 and is about 300 m thickrepresenting 1.5 × 105 km3 of preserved sediment, but this was depos-ited over about 5 Ma. Mass-balance calculations of Hampson et al.(2014) suggest significant along-strike loss of sediment.

Studies of the Campanian Ericson Formation inWyoming show sim-ilar scale channels, averaging several meters deep, with the deepestchannels up to about 15 m (Martinsen et al., 1999). Pranter et al.(2009) compiled sand-body dimensional data for the Williams ForkFormation and showed the thickest single-story sand bodies to be upto 9 m deep, suggesting similarity to the Blackhawk Formation. To

Fig. 29. A. Cross section of clinoform strata in the Campanian of Wyoming. Clinothem 9 is highlighted on cross section and shown on isopach map (B). Fluvial channels in the LanceFormation feed deep-water sandstones in the Lewis Shale (from Carvajal and Steel, 2009).


date there have been no attempts to estimate paleodischarge orpaleodrainage for the Ericson or William's Fork systems.

By the Maastrichtian, a new drainage divide associated with theLaramide orogeny emerged that closely paralleled the Canada-USA bor-der (Fig. 8B). Rivers of the Lance Formation flowed largely south,draining off this divide (Carvajal and Steel, 2009). A distinct fluvial,shelf-slope and linked deep-water system developed for the first timein the Washakie and Great Divide basins of southern Wyoming duringthe Maastrichtian and the basin reached up to 500 m deep (Fig. 29).Carvajal and Steel (2009) undertook detailed sequence stratigraphicanalysis of the Lance, Fox Hills and Lewis Shale, which represent thefluvial, shoreline, and deep-water components respectively. Fourteenseparate clinothem units were correlated and mapped (Fig. 29).The total clinoform set was deposited in 1.8 Ma, suggesting eachclinothem unit represents about 10,000 years. These clinothemunits typically contain several parasequences deposited in a fewthousand years, suggesting rates typical of modern systems (Walshand Nittrouer, 2009). The clinothems were also grouped into twodistinct stages (Carvajal and Steel, 2009, 2012). Stage 1 is dominatedby aggradational stacking and correlated to a time of higher subsi-dence. Stage 2 is dominantly progradational and correlates to atime of increased tectonic uplift which is inferred to have generatedhigher sediment supply.

Fluvial trunk channels in the Lance Formation were about 7 m deep.Associated sandy deep-water fans in the Lewis Shale were quite small,covering an area of 500–750 km2. Fan deposits reach thicknesses of upto 100 m, with up to 60 m of sandstone (Fig. 29). Carvajal and Steel(2012) estimated sandstone volumes of up to 66 km3. Total clinothem

volumes of 630 km3 were calculated by Carvajal and Steel (2012), butrepresent minimal estimates, as the proximal portions of clinothemswere not included. Drainage areas were estimated to range from 5 to20 × 103 km2 during stage 1, and expanded to 20–35 × 103 km2, duringstage 2. These drainages are comparable to the Ferron systemsdiscussed in Section 7.6, and river depths are correspondingly similar(e.g., 5–7 m deep). They also calculated sediment yields ranging from200 to 2000 t/km2 y and sediment loads ranging from 4 to 16 Mt./y(Mt = megatons). Carvajal and Steel (2012) also used the BQARTmodel (Syvitski and Milliman, 2007) to estimate that the relief ofthe source area changed from lowland (100–500 m) to upland (500–1000 m) in stage 1, to mountainous (1000–3000 m) in stage 2. Thecalculated sediment loads indicate modest rivers, similar in scale tothose in Southeast Asian islands (e.g., New Guinea, Sumatra, Borneo),and consistent with small drainage basins. The higher sediment yieldsare also comparable with analogous small mountainous riversand could also have generated frequent hyperpycnal flows (Millimanand Syvitski, 1992). Soyinka and Slatt (2008) identified abundanthyperpycnites and turbidites in the Lewis Shale in Wyoming, whichwere fed by these rivers. Petter et al. (2013) present a simpler alternatemethod for estimating sediment paleoflux using stratigraphic criteria.Clinoform dimensions and stacking patterns were used to measure thedegree of aggradation and progradation, whichwas equated to changesin sediment discharge rates. Their estimates of sediment discharge forthe Lance-Fox Hills-Lewis system were 8.8Mt/y, well within the4-16Mt/y range estimated by Carvajal and Steel (2012).

The fluvial Lance sandstones form significant tight gas reservoirs inthe Jonah and Pinedale Field in Wyoming (e.g., Longman et al., 2015


and references therein), and the degree of sandstone amalgamation aswell as scale of channels are important in controlling recovery. Thesubmarine fan sandstones in the Lewis host gas reservoirs in theGreen River Basin, Wyoming (Muller and Wirnkar, 2004) and thesouthern extension of the Lewis shale forms important gas reservoirsin Colorado and New Mexico (Dubiel, 2007). The quantitative S2Sapproach suggests that the moderate-paleodischarge Lance paleoriversdirectly fed the interpreted hyperpycnites and turbidites of the deep-water Lewis Shale (Soyinka and Slatt, 2008) and illustrates the valueof this approach in understanding the nature of linked fluvial to deep-water depositional systems, with implications for hydrocarbon discov-ery and recovery.

8. Conclusion and future work

This paper shows how quantitative source-to-sink analysis can beused to make estimates regarding the paleodischarge of ancient riversystems and in turn predictions about the volume and characteristicsof deposits in downstream sinks, aswell as the size, relief, slope, and cli-mate regime of upstream areas. This approach requires analysis of thescale (e.g., depth), and/or sedimentology (e.g., grain size, sedimentarystructures) of rivers to determine their velocity and discharge. For ex-ample, documentation of the relatively small nature of trunk rivers inthe Ferron, Dunvegan and Lance, indicate they could have routinely pro-duced hyperpycnal flows during large magnitude flood events (Mulderand Syvitski, 1995). Examination of associated downdip prodelta anddeep-water facies show locally abundant hyperpycnites and turbidites,likely fed by these rivers (Bhattacharya andMacEachern, 2009; Soyinkaand Slatt, 2008). These heterolithic facies are now being exploited asthin-bedded reservoirs that fringe the tapped-out conventionalsandstone play in the Dunvegan Formation, and form deep-water gasreservoirs in the Lewis Shale.

In steep gradient S2S systems (Slope ~10−3), such as the Frontierand Cardium formations, backwater lengths were short (b10 km),which allows gravel to reach the coast, and these shallow marineconglomerates form important oil and gas reservoirs. In lower gradientsystems (S = 10−4–10−5), such as the McMurray and Dunveganbackwater lengths are higher (N100 km), and gravel phases are con-fined to the inland alluvial system, and coastal systems are sandy,precluding the development of coarser-grained reservoirs. TheLower Cretaceous McMurray Formation, was a continental-scale,low-gradient river system enabling development of an extensivetidal backwater zone. Abundant tidal features and lack of gravel-grade sediment support this prediction. Because channel depositsare large, internal heterogeneity, and especially tidally-influencedpoint bar mud drapes, can be mapped and modeled explicitly, whichis important in managing both steam floods in subsurface and miningoperations at the surface.

Dimensions and paleo-discharge estimates of Mesozoic paleoriverswas integrated with provenance analysis and regional paleotectonicmaps to document the evolution of drainage areas in North Americathroughout the Mesozoic. These show three major phases. The firstphase as associated with continental-scale rivers draining southernand easterly (Appalachian source areas) in the Triassic through Albian.The second phase is marked by a transition to smaller drainage basinsin the Cenomanian-Campanian, primarily draining the Western Cordil-lera, and reflecting the development of the Sevier Orogeny, whichformed an extensive foreland basin along the thrust front that wasflooded by the Cretaceous Interior Seaway. The third phase shows a re-turn to a continental-scale drainage in the Paleocene as the Seawayretreated and the foreland basin rebounded.

During the Cenomanian to Maastrichtian, river and delta systems,such as the Dunvegan, Ferron and Lance, were relatively small, allowingclosure of S2S sediment budgets. Comparison of predicted sedimenttransfer to mapped sediment volumes, such as in the Dunvegan, andLance-Fox Hills-Lewis systems, show remarkable concordance, despite

the uncertainties and errors that may be associated with the S2Sapproach.

Although it was not the focus of this paper, future regional strati-graphic S2S analysis must be integrated with geodynamic analysis tobetter link tectonic processes with the sedimentary record. This is criti-cal to determine the relative role of tectonic versus climatic processes informing the stratigraphic record. To that end, integration of detailedanalysis of floodplains and their paleosols may help to better determinepaleoclimate variability. A remaining challenge is high-resolution chro-nometry. The sequence stratigraphic approach is critical in that it en-ables subdivisions of formations into smallest-scale, mappable geneticunits, identified by integrating environmental lithofacies and boundingsurfaces, (i.e., parasequences, clinothems) fromwhich estimates of sed-iment volumes associatedwith a specific paleo-river can bemade. Theseparasequences, however, are typically finer-scale than the precision ofisotope chronometers, and assigning absolute durations to these unitsinvolves significant uncertainty. Recent higher-precision methods ofSinger et al. (2015) provide immense promise in determination ofhigh-resolution astrochronological (Milankovitch) cycles. These im-proved chronostratigraphic tools could offer much insights, but despitethis, in a number of examples discussed in this paper, there is remark-able conformity between predictions of sediment transfer and mappedsedimentary volumes, indicating that the S2S approach is robust andcan be applied to deep-time systems.Morever, such effortsmay providenewknowledge that can be economically beneficial to oil and gas explo-ration and production.

Acknowledgements

This paper was written with the generous support of the McMasterUniversity/University of HoustonQuantitative Sedimentology Laborato-ries, including current funding from BP, and Inpex and a NaturalSciences and Engineering Research Council of Canada Discovery Grantto Bhattacharya. We also acknowledge former support from Anadarko,BHP Billiton, Chevron, Ecopetrol, ExxonMobil, Pioneer Natural Re-sources, and Shell. We thank Steve Hubbard and Christian Carvajal fortheir thorough reviews, and Special Edition editor J.P. Walsh for hisinsightful comments and meticulous editing.

References

Aalto, R., Lauer, J., Dietrich, W.E., 2008. Spatial and temporal dynamics of sediment accu-mulation and exchange along Strickland River floodplains (Papua New Guinea) overdecadal-to-centennial timescales. J. Geophys. Res. Earth Surf. 113 (F1). http://dx.doi.org/10.1029/2006JF000627.

Adams, M.M., Bhattacharya, J.P., 2005. 2005. No change in fluvial style across a sequenceboundary, Cretaceous Blackhawk and castlegate formations of central Utah, U.S.A.J. Sediment. Res. 75, 1040–1053.

Adams, R.L., Carr, J.P., 2010. Regional Depositional Systems of the Woodbine, Eagle Ford,and Tuscaloosa of the US Gulf Coast. Gulf Coast Association of Geological SocietiesTransactions 60, pp. 3–27.

Ager, D.V., 1973. The Nature of the Stratigraphical Record. Macmillan, London.Ahmed, S., Bhattacharya, J.P., Garza, D.E., Li, Y., 2014. Facies architecture and stratigraphic

evolution of a river-dominated delta front, Turonian Ferron sandstone, Utah. JSR 84(2), 97–121.

Akyuz, I., Warny, S., Famubode, O., Bhattacharya, J.P., 2015. Palynology of the TuronianFerron-Notom Sandstone, Utah: identification of marine flooding surfaces andMilankovitch cycles in subtropical, ever-wet, paralic to non-marine paleoenvironments.Palynology http://dx.doi.org/10.1080/01916122.2015.1014525.

Allen, P.A., Allen, J.R., 2013. Basin Analysis: Principles and Application to Petroleum PlayAssessment. John Wiley & Sons.

Ambrose, W.A., Hentz, T.F., Bonnaffe, F., Loucks, R.G., Brown Jr., L.F., Wang, F.P., Potter, E.C.,2009. Sequence-stratigraphic controls on complex reservoir architecture of highstandfluvial-dominated deltaic and lowstand valley-fill deposits in the Upper Cretaceous(Cenomanian) Woodbine Group, East Texas field: regional and local perspectives.AAPG Bull. 93 (2), 231–269.

Anderson, J., Rodrigues, A., Abdulah, K., Fillon, R.H., Banfield, L., Mckeown, H., Wellner, J.,2004. Late Quaternary stratigraphic evolution of the northern Gulf of Mexico margin:A synthesis. In: Anderson, J.B., Fillon, R.H. (Eds.), Late Quaternary Stratigraphic Evolu-tion of the Northern Gulf of Mexico Margin. SEPM special Publication 79, pp. 1–23.

Arnott, R.W.C., 1991. The Carrot Creek” K” pool, Cardium Formation, Alberta; a conglom-eratic reservoir related to a wave-reworked distributary mouth-bar complex. Bull.Can. Petrol. Geol. 39 (1), 43–53.

http://dx.doi.org/10.1029/2006JF000627

http://refhub.elsevier.com/S0012-8252(15)30059-3/rf0010










http://dx.doi.org/10.1080/01916122.2015.1014525














Arnott, R.W.C., 1992. The role of fluvial processes during deposition of the (Cardium)Carrot Creek/Cyn-Pem conglomerates. Bull. Can. Petrol. Geol. 40 (4), 356–362.

Barrell, J., 1917. Rhythms and the measurements of geologic time. Bull. Geol. Soc. Am. 28,745–904.

Beaumont, C., Fullsack, P., Hamilton, J., 1992. Erosional Control of Active CompressionalOrogens. In: McClay, K.R. (Ed.), Thrust tectonics, pp. 1–18.

Bentley Sr., S.J., 2003. Wave-current Dispersal of Fine-grained Fluvial sediments AcrossContinental Shelves: the Significance of Hyperpycnal Plumes. In: Scott, E.D., Bouma,A.H., Bryant, W.R. (Eds.), Siltstones, Mudstones and Shales: Depositional Processesand Characteristics: SEPM CD, pp. 35–48.

Benyon, C., Leier, A.L., Leckie, R.M., Webb, A., Hubbard, S.M., Gehrels, G.E., 2014. Prove-nance of the Cretacous Athabasca oil sands, Canada: implications for continental-scale sediment transport. J. Sediment. Res. 84, 136–143.

Bergman, K.M., Snedden, J.W., 1999. Isolated shallowmarine sand bodies: sequence strat-igraphic analysis and sedimentologic interpretation. In: Bergman, K.M., Snedden, J.(Eds.), Isolated Shallow Marine Sand Bodies: Sequence Stratigraphic Analysis andSedimentologic Interpretation. SEPM Special. Publication 64 (362 pp.).

Bergman, K.M., Walker, R.G., 1987. The importance of sea-level fluctuations in theformation of linear conglomerate bodies; Carrot Creek Member of CardiumFormation, Cretaceous Western Interior Seaway, Alberta, Canada. J. Sediment.Res. 57 (4), 651–665.

Bernet, M., van der Beek, P., Pik, R., Huyghe, P., Mugnier, J.-L., Labrin, E., Szulc, A., 2006.Miocene to recent exhumation of the central Himalaya determined from combineddetrital zircon fission-track and U/Pb analysis of Siwalik sediments, western Nepal.Basin Res. 18, 393–412.

Bhattacharya, J.P., 1993. The expression and interpretation of marine flooding surfacesand erosional surfaces in core; examples from the Upper Cretaceous Dunvegan For-mation in the Alberta foreland basin. In: Summerhayes, C.P., Posamentier, H.W.(Eds.), Sequence stratigraphy and facies associations. IAS special publication 18,pp. 125–160.

Bhattacharya, J.P. 1994. Cretaceous Dunvegan Formation strata of the Western CanadaSedimentary Basin. in: Mossop, G.D., and Shetsen, I., (comps.) Geological Atlas ofthe Western Canada Sedimentary Basin. Canadian Society of Petroleum Geologistsand Alberta Research Council, p. 365–373.

Bhattacharya, J.P., 2010. In: Dalrymple, R.G., James, N.P. (Eds.), Facies Models 4. GeologicalAssociation of Canada. Geotext 6, pp. 233–264.

Bhattacharya, J.P., 2011. Practical problems in the application of the sequence strat-igraphic method and key surfaces: integrating observations from ancient fluvial–deltaic wedges with quaternary and modelling studies. Sedimentology 58,120–169.

Bhattacharya, J.P., Giosan, L., 2003. Wave-influenced deltas: geomorphological implica-tions for facies reconstruction. Sedimentology 50, 187–210.

Bhattacharya, J.P., MacEachern, J.A., 2009. Hyperpycnal Rivers and Prodeltaic Shelves inthe Cretaceous Seaway of North America. J. Sediment. Res. 79, 184–209.

Bhattacharya, J.P., Tye, R.S., 2004. Searching for modern ferron analogs and application tosubsurface interpretation. In: Chidsey Jr., T.C., Adams, R.D., Morris, T.H. (Eds.), TheFluvial-Deltaic Ferron Sandstone: Regional to Wellbore-Scale Outcrop Analog Studiesand Application to Reservoir Modeling. AAPG Studies in Geology 50, pp. 39–57.

Bhattacharya, J., Walker, R.G., 1991. Allostratigraphic subdivision of the upper cretaceousdunvegan, Shaftesbury, and Kaskapau Formations in the subsurface of northwesternAlberta. Bull. Can. Petrol. Geol. 39, 145–164.

Bhattacharya, J.P., Willis, B.J., 2001. Lowstand Deltas in the Frontier Formation, PowderRiver Basin, Wyoming: Implications for sequence stratigraphic models, U.S.A. AAPGBull. 85, 261–294.

Bhattacharyya, P., Bhattacharya, J.P., Khan, S.D., 2015. Paleo-Channel Reconstruction andGrain Size Variability in Fluvial Deposits, Ferron Sandstone, Notom Delta, HanksvilleUtah. Sedimentary Geology 325, 17–25.

Blair, T.C., McPherson, J.G., 1994. Alluvial fans and their natural distinction from riversbased on morphology, hydraulic processes, sedimentary processes, and facies assem-blages. J. Sediment. Res. 64, 450–489.

Blakey, R., 2015. Library of Paleogeography, Colorado Plateau Geosystems™, Arizona, USA.(website: http://cpgeosystems.com/index.html).

Blum, M.D., 2015. Scaling relationships between fluvial channel fills, channel-belt sandbodies and drainage basins, with implications for the mannville group, albertaforeland basin. AAPG Annual Convention Program with Abstracts.

Blum, M., Pecha, M., 2014. Mid-Cretaceous to Paleocene North American drainage reorga-nization from detrital zircons. Geology 42, 607–610.

Blum, M.D., Törnqvist, T.E., 2000. Fluvial responses to climate and sea-level change: areview and look forward. Sedimentology 47, 2–48.

Blum, M., Martin, B.J., Milliken, K., Garvin, M., 2013. Paleovalley systems: insights fromquaternary analogs and experiments. Earth Sci. Rev. 116, 128–169.

Boyd, R., Dalrymple, R.W., Zaitlin, B.A., 2006. Estuarine and incised valley facies models.In: Walker, R.G., Posamentier, H. (Eds.), Facies Models Revisited. SEPM, SpecialPublication 84, pp. 171–235.

Bridge, J.S., 2003. Rivers and Floodplains. Blackwell.Bridge, J.S., Tye, R.S., 2000. Interpreting the dimensions of ancient fluvial channel bars,

channels, and channel belts from wireline-logs and cores. AAPG Bull. 84 (8),1205–1228.

Brownlie, W.R., 1983. Flow depth in sand-bed channels. J. Hydraul. Eng. 109, 959–990.Buechmann, D., 2013. Provenance, Detrital Zircon U–Pb Geochronology, and Tectonic Sig-

nificance of Middle Cretaceous Sandstones from the Alberta Foreland Basin. Universi-ty of Houston (Master of Science Thesis, 96 pp.).

Campbell, I.H., Reiners, P.W., Allen, C.M., Nicolescu, S., Upadhyay, R., 2005. He–Pb doubledating of detrital zircons from the Ganges and Indus rivers: implication for quantify-ing sediment recycling and provenance studies. Earth Planet. Sci. Lett. 237 (3),402–432.

Carter, L., Orpin, A.R., Kuehl, S.A., 2010. From mountain source to ocean sink–the passageof sediment across an active margin, Waipaoa Sedimentary System, New Zealand.Mar. Geol. 270 (1), 1–10.

Carvajal, C., Steel, R., 2009. Shelf-edge architecture and bypass of sand to deep water: in-fluence of shelf-edge processes, sea level, and sediment supply. J. Sediment. Res. 79(9), 652–672.

Carvajal, C., Steel, R., 2012. Source-to-sink sediment volumes within a tectono-strati-graphic model for a laramide shelf-to-deep-water basin: methods and results. In:Busby, C., Perez, A.A. (Eds.), Tectonics of Sedimentary Basins: Recent Advances,pp. 131–151.

Carvajal, C., Steel, R., Petter, A., 2009. Sediment supply: the main driver of shelf-margingrowth. Earth Sci. Rev. 96 (4), 221–248.

Catuneanu, O., 2002. Sequence stratigraphy of clastic systems: concepts, merits, and pit-falls. J. Afr. Earth Sci. 35 (1), 1–43.

Catuneanu, O., Abreu, V., Bhattacharya, J.P., Blum, M.D., Dalrymple, R.W., Eriksson, P.G.,Fielding, C.R., Fisher, W.L., Galloway, W.E., Gibling, M.R., Giles, K.A., Holbrook, J.M.,Jordan, R., Kendall, C.G.S.C., Macurda, B., Martinsen, O.J., Miall, A.D., Neal, J.E.,Nummedal, D., Pomar, L., Posamentier, H.W., Pratt, B.R., Sarg, J.F., Shanley, K.W.,Steel, R.J., Strasser, A., Tucker, M.E., Winker, C., 2009. Toward the standardization ofsequence stratigraphy. Earth Sci. Rev. 92, 1–33.

Chidsey, T.C., Adams, R.D., Morris, T.H., 2004. Regional to Wellbore Analog for Fluvial-deltaic Reservoir Modeling: the Ferron Sandstone of Utah. AAPG Studies in Geology50 (568 pp.).

Copeland, P., Harrison, T.M., 1990. Episodic rapid uplift in the Himalaya revealed by 40Ar/39Ar analysis of detrital K-feldspar and muscovite, Bengal Fan. Geology 18, 354–357.

Corbeanu, R.M., Wizevich, M.C., Bhattacharya, J.P., Zeng, X., McMechan, G.A., 2004. Three-dimensional architecture of ancient lower delta-plain point bars using groundpenetrating-radar, Cretaceous Ferron sandstone, Utah. In: Chidsey Jr., T.C., Adams,R.D., Morris, T.H. (Eds.), The Fluvial-Deltaic Ferron Sandstone: Regional toWellbore-Scale Outcrop Analog Studies and Application to Reservoir Modeling.AAPG Studies in Geology 50, pp. 427–449.

Cotter, E., 1975. Deltaic deposits in the Upper Cretaceous Ferron Sandstone, Utah. In:Broussard, M.L. (Ed.), Deltas, models for exploration: Houston Geological Society,pp. 471–484.

Covault, J.A., Graham, S.A., 2010. Submarine fans at all sea-level stands: tectono-morphologic and climatic controls on terrigenous sediment delivery to the deepsea. Geology 38 (10), 939–942.

Covault, J.A., Normark, W.R., Romans, B.W., Graham, S.A., 2007. Highstand fans in theCalifornia borderland: the overlooked deep-water depositional systems. Geology 35(9), 783–786.

Creaney, S., Allan, J., Cole, K. S., Fowler, M. G., Brooks, P. W., Osadetz, K. G., and Riediger, C.L., 1994. Petroleum generation and migration in the Western Canada SedimentaryBasin. in: Mossop, G.D. and Shetsen, I., (Comps.), Geological Atlas of the WesternCanada Sedimentary Basin, Canadian Society of Petroleum Geologists and Alberta Re-search Council, pp. 455–468.

Critelli, S., Ingersoll, R.V., 1995. Interpretation of neovolcanic versus palaeovolcanic sandgrains: an example from Miocene deep-marine sandstone of the Topanga group(Southern California). Sedimentology 42 (5), 783–804.

Davidson, S.K., Hartley, A.J., 2010. A quantitative approach to linking drainage area withdistributive fluvial system area in dryland, endorheic basin settings. Geomorphology180 (181), 82–95.

Davidson, S.K., North, C.P., 2009. Geomorphological regional curves for prediction ofdrainage area and screening modern analogues for rivers in the rock record.J. Sediment. Res. 79 (10), 773–792.

Davies, N.S., Gibling, M.R., 2010. Cambrian to Devonian evolution of alluvial systems: thesedimentological impact of the earliest land plants. Earth Sci. Rev. 98 (3), 171–200.

Dawson, F. M., Evans, C. G., Marsh, R., & Richardson, R., 1994. Uppermost Cretaceous andTertiary strata of the Western Canada sedimentary basin. in: Mossop, G.D. andShetsen, I., (comps.). Geological atlas of the Western Canadian Sedimentary Basin,Canadian Society of Petroleum Geologists and Alberta Research Council, p.387–406.

DeCelles, P.G., 2004. Late Jurassic to Eocene evolution of the Cordilleran thrust belt andforeland basin system, western USA. Am. J. Sci. 304 (2), 105–168.

Deveugle, P.E., Jackson, M.D., Hampson, G.J., Stewart, J., Clough, M.D., Ehighebolo, T.,Farrell, M.E., Calver, C.S., Miller, J.K., 2014. A comparative study of reservoir modelingtechniques and their impact on predicted performance of fluvial-dominated deltaicreservoirs. AAPG Bull. 98 (4), 729–763.

Dickin, A.P., 2005. Radiogenic Isotope Geology. Cambridge University Press (492 pp.).Dickinson, W.R., 1985. Interpreting provenance relations from detrital modes of sand-

stones. In: Zuffa, G.G. (Ed.), Provenance of arenites. Springer, Netherlands,pp. 333–361.

Dickinson, W.R., 2003. The place and power of myth in geoscience: an associate editor'sperspective. Am. J. Sci. 303 (9), 856–864.

Dickinson, W.R., Gehrels, G.E., 2003. U–Pb ages of detrital zircons from Permian and Juras-sic eolian sandstones of the Colorado Plateau, USA: paleogeographic implications.Sediment. Geol. 163 (1), 29–66.

Dickinson, W.R., Gehrels, G.E., 2008. U–Pb ages of detrital zircons in relation to paleogeog-raphy: Triassic paleodrainage networks and sediment dispersal across southwestLaurentia. J. Sediment. Res. 78 (12), 745–764.

Dickinson, W.R., Gehrels, G.E., 2009. U–Pb ages of detrital zircons in Jurassic eolian and as-sociated sandstones of the Colorado plateau: evidence for transcontinental dispersaland intraregional recycling of sediment. Geol. Soc. Am. Bull. 121 (3–4), 408–433.

Dickinson, W.R., Suczek, C.A., 1979. Plate tectonics and sandstone compositions. AAPGBull. 63 (12), 2164–2182.

Dickinson, W.R., Gehrels, G.E., Marzolf, J.E., 2010. Detrital zircons from fluvial Jurassic stra-ta of the Michigan basin: implications for the transcontinental Jurassic paleoriver hy-pothesis. Geology 38 (6), 499–502.
























































http://cpgeosystems.com/index.html


































































































Dolson, J., Muller, D., Evetts, M.J., Stein, J.A., 1991. Regional paleotopographic trends andproduction, muddy sandstone (Lower Cretaceous), Central and Northern RockyMountains (1). AAPG Bull. 75 (3), 409–435.

Donselaar, M.E., Overeem, I., 2008. Connectivity of fluvial point-bar deposits: an examplefrom theMiocene Huesca fluvial fan, Ebro basin, Spain. AAPG Bull. 92 (9), 1109–1129.

Draut, A.E., Kineke, G.C., Velasco, D.W., Allison, M.A., Prime, R.J., 2005. Influence of theAtchafalaya river on recent evolution of the Chenier-plain inner continental shelf,northern gulf of Mexico. Cont. Shelf Res. 25 (1), 91–112.

Driscoll, N., Nittrouer, C., 2000. Source-to-sink studies. Margins 11, 14.Dubiel, R.F., 1992. Sedimentology and Depositional History of the Upper Triassic Chinle

Formation in the Uinta, Piceance, and Eagle Basins, Northwestern Colorado andNortheastern Utah.

Dubiel, R.F., 1994. Triassic deposystems, paleogeography, and paleoclimate of thewestern interior. In: Caputo, M.V., Peterson, J.A., Franczyk, K.J. (Eds.), MesozoicSystems of the Rocky Mountain Region, USA. Rocky Mountain Section SEPM,Denver, pp. 133–168.

Dubiel, R.F., 2007. Sequence stratigraphy applied to the oil and gas assessment of theLewis shale total petroleum system, San Juan basin, Colorado and New Mexico. Mt.Geol. 44 (4), 155–173.

Eaton, J.G., 1991. Biostratigraphic framework for the upper cretaceous rocks of thekaiparowits plateau, southern Utah. In: Nations, J.D., Eaton, J.G. (Eds.), Stratigraphy,Depositional Environments, and Sedimentary Tectonics of the Western Margin. Cre-taceous Western Interior Seaway: Geological Society of America Special Paper 260,pp. 47–63.

Edwards, D.E., Barclay, J.E., Gibson, D.W., Kvill, G.E., and Halton, E., 1994. Triassic Strata ofthe Western Canada Sedimentary Basin. in: Mossop, G.D., and Shetsen, I., (Comp.),Geological Atlas of theWestern Canada sedimentary basin, Canadian Society of Petro-leum Geologists and Alberta Research Council. pp. 257–275.

Fielding, C.R., 2010. Planform and facies variability in asymmetric deltas: facies analysisand depositional architecture of the Turonian Ferron sandstone in the WesternHenry mountains, south-central Utah, USA. J. Sediment. Res. 80 (5), 455–479.

Fielding, C.R., 2011. Foreland basin structural growth recorded in the Turonian Ferronsandstone of the Western Interior Seaway Basin, USA. Geology 39 (12), 1107–1110.

Fielding, C.R., 2015. Anatomy of falling-stage deltas in the Turonian Ferron sandstone ofthe western Henry mountains syncline, Utah: growth faults, slope failures andmass transport complexes. Sedimentology 62 (1), 1–26.

Finzel, E.S., 2014. Detrital zircons from cretaceous midcontinent strata reveal an Appala-chian mountains–cordilleran foreland basin connection. Lithosphere 6, 378–382.

Flood, Y.S., Hampson, G.J., 2015. Quantitative analysis of the dimensions and distributionof channelized fluvial sandbodies within a large outcrop dataset: Upper CretaceousBlackhawk formation, Wasatch plateau, central Utah, USA. J. Sediment. Res. 85 (4),315–336.

Francis, J.M., Daniell, J.J., Droxler, A.W., Dickens, G.R., Bentley, S.J., Peterson, L.C., Opdyke,B.N., Beaufort, L., 2008. Deep water geomorphology of the mixed siliciclastic-carbonate system, gulf of Papua. J. Geophys. Res. Earth Surf. 113 (F1). http://dx.doi.org/10.1029/2007JF000851.

Franzinelli, E., Potter, P.E., 1983. Petrology, chemistry, and texture of modern river sands,amazon river system. J. Geol. 91, 23–39.

Fustic, M., Hubbard, S.M., Spencer, R., Smith, D.G., Leckie, D.A., Bennett, B., Larter, S., 2012.Recognition of down-valley translation in tidally influenced meandering fluvial de-posits, Athabasca oil sands (Cretaceous), Alberta, Canada. Mar. Pet. Geol. 29 (1),219–232.

Fustic, M., Cadiou, D., Thurston, D., Al-Dliwe, A., Leckie, D.A., 2013. Reservoir model-ing by constraining stochastic simulation to determin- istically interpretedthree-dimensional geobodies: case study from Lower Cretaceous McMurray for-mation, long lake steam-assisted gravity drainage project, Northeast Alberta,Canada. In: Hein, F.J., Leckie, D.A., Larter, S., Suter, J.R. (Eds.), Heavy-oil and Oil-sand Petroleum Systems in Alberta and Beyond. AAPG Studies in Geology 64,pp. 565–604.

Galloway, W.E., 1982. Depositional architecture of Cenozoic gulf coastal plain fluvial sys-tems. In: Ethridge, F.G., Flores, R.M. (Eds.), Recent and ancient nonmarine deposition-al environments: models for exploration. SEPM Special Publication 31, pp. 127–155.

Galloway, W.E., 1986. Growth faults and fault-related structures of prograding terrige-nous clastic continental margins. Gulf Coast Association of Geological Societies Trans-actions 36, pp. 121–128.

Galloway, W.E., Whiteaker, T.L., Ganey-Curry, P., 2011. History of Cenozoic northAmerican drainage basin evolution, sediment yield, and accumulation in the gulf ofMexico basin. Geosphere 7 (4), 938–973.

Gani, M.R., Bhattacharya, J.P., 2007. Basic building blocks and process variability of a cre-taceous delta: internal facies architecture reveals a more dynamic interaction of river,Wave, and Tidal Processes Than Is Indicated by External Shape. J. Sediment. Res. 77,284–302.

Gardner, M., 1995. Tectonic and eustatic controls on the stratal architecture of mid-Cretaceous stratigraphic sequences, central western interior foreland basin of NorthAmerica. In: Dorobek, S.L., Ross, G.M. (Eds.), Stratigraphic Evolution of Foreland Ba-sins: SEPM Special Publication 52, pp. 243–281.

Garrison Jr., J.R., Van den Bergh, T.C.V., 2004. High-resolution depositional sequence stra-tigraphy of the upper Ferron sandstone last chance delta: an application of coal-zonestratigraphy. In: Chidsey Jr., T.C., Adams, R.D., Morris, T.H. (Eds.), Regional toWellboreAnalog for Fluvial-Deltaic Reservoir Modeling: The Ferron Sandstone of Utah. AAPGStudies in Geology 50, pp. 125–192.

Garrison Jr., J.R., van denBergh, T.C.V., 2006. Effects of sedimentation rate, rate of relative risein sea level, and duration of sea-level cycle on the filling of incised valleys: examples offilled and “overfilled” incised valleys from the Upper Ferron sandstone, last chancedelta, east-central Utah, U.S.A. In: Dalrymple, R.W., Leckie, D.A., Tillman, R.W. (Eds.), In-cised Valleys in Time and Space. SEPM Special Publication 85, pp. 239–279.

Gehrels, G.E., Dickinson, W.R., Ross, G.M., Stewart, J.H., Howell, D.G., 1995. Detrital zirconreference for Cambrian to Triassic miogeoclinal strata of western North America.Geology 23 (9), 831–834.

Gibling, M.R., Davies, N.S., 2012. Palaeozoic landscapes shaped by plant evolution. Nat.Geosci. 5 (2), 99–105.

Gibson, D.W., Barclay, J.E., 1989. Middle Absaroka sequence the Triassic stable craton. In:Ricketts, B.D. (Ed.), Western Canada Sedimentary Basin, A Case History. CanadianSociety of Petroleum Geologists, pp. 219–231.

Gong, C., Wang, Y., Steel, R.J., Olariu, C.X.Q., Liu, X., Zhao, Q., 2015. Growth styles of shelf-margin clinoforms; prediction of sand- and sediment-budget partitioning into andacross the shelf. J. Sediment. Res. 85 (3), 209–229.

Goodbred, S.L., Kuehl, S.A., 1999. Holocene and modern sediment budgets for the Ganges-Brahmaputra river system: evidence for highstand dispersal to flood-plain, shelf, anddeep-sea depocenters. Geology 27 (6), 559–562.

Goodbred Jr., S.L., Kuehl, S.A., 2000. The significance of large sediment supply, active tec-tonism, and eustasy on margin sequence development: late quaternary stratigraphyand evolution of the Ganges–Brahmaputra delta. Sediment. Geol. 133 (3), 227–248.

Grantz, A., May, S.D., Hart, P.E., 1990. Geology of the Arctic continental margin of Alaska.In: Grantz, A.A., Johnson, L., Sweeney, J.F. (Eds.), The Geology of North America, Geo-logical Society of America 50, pp. 257–288.

Guillocheau, F., Rouby, D., Robin, C., Helm, C., Rolland, N., Le Carlier de Veslud, C., Braun, J.,2012. Quantification and causes of the terrigeneous sediment budget at the scale of acontinental margin: a new method applied to the Namibia–South Africa margin.Basin Res. 24 (1), 3–30.

Hajek, E.A., Heller, P.L., 2012. Flow-depth scaling in alluvial architecture and nonmarinesequence stratigraphy: example from the castlegate sandstone, central Utah, USA.J. Sediment. Res. 82 (2), 121–130.

Hampson, G.J., Gani, M.R., Sahoo, H., Rittersbacher, A., Irfan, N., Ranson, A., Jewell, T.O.,Gani, N.D., Howell, J.A., Buckley, S.J., Bracken, B., 2012. Controls on large-scale patternsof fluvial sandbody distribution in alluvial to coastal plain strata: Upper CretaceousBlackhawk formation, Wasatch plateau, Central Utah, USA. Sedimentology 59 (7),2226–2258.

Hampson, G.J., Jewell, T.O., Irfan, N., Gani, M.R., Bracken, B., 2013. Modest change in fluvialstyle with varying accommodation in regressive alluvial-to-coastal-plain wedge:Upper Cretaceous Blackhawk formation, Wasatch plateau, central Utah, USA.J. Sediment. Res. 83 (2), 145–169.

Hampson, G.J., Duller, R.A., Petter, A.L., Robinson, R.A.J., Allen, P.A., 2014. Mass-balanceconstraints on stratigraphic interpretation of linked alluvial-coastal-shelfal depositsfrom source-to-sink: example from cretaceous western interior basin, Utah andColorado, USA. J. Sediment. Res. 84, 935–960.

Hansley, P.L., 1986. Relationship of detrital, nonopaque heavy minerals to diagenesis andprovenance of the Morrison formation, southwestern San Juan basin. New MexicoAAPG Bulletin 179, pp. 257–276.

Haq, B.U., Hardenbol, J., Vail, P.R., 1987. Chronology of fluctuating sea levels since theTriassic. Science 235 (4793), 1156–1167.

Hart, B.S., Plint, A.G., 2003. Stratigraphy and sedimentology of shoreface and fluvial con-glomerates: insights from the Cardium formation in NW Alberta and adjacent BritishColumbia. Bull. Can. Petrol. Geol. 51 (4), 437–464.

Hartley, A.J., Weissmann, G.S., Nichols, G.J., Warwick, G.L., 2010. Large distributive fluvialsystems: characteristics, distribution, and controls on development. J. Sediment. Res.80 (2), 167–183.

Hayes, B. J. R., Christopher, J. E., Rosenthal, L., Los, G., McKercher, B., Minken, D., Tremblay,Y.M., Fennell, and Smith, D. G., 1994. Cretaceous Mannville Group of the WesternCanada Sedimentary Basin. In: Mossop, G.D., and Shetsen, I., (Comps.), GeologicalAtlas of the Western Canada sedimentary basin, Canadian Society of PetroleumGeologists and Alberta Research Council. pp. 317–334.

Helland-Hansen, W., Martinsen, O.J., 1996. Shoreline trajectories and sequences: descrip-tion of variable depositional-dip scenarios. J. Sediment. Res. 66, 670–688.

Hill, P.S., Fox, J.M., Crockett, J.S., Curran, K.J., Friedrichs, C.T., Geyer, W.R., Milligan,T.G., Ogston, A.S., Puig, P., Scully, M.E., Traykovski, P.A., Wheatcroft, R.A., 2007.Sediment delivery to the seabed on continental margins. In: Nittrouer, C.A.,Austin, J.A., Field, M.E., Kravitz, J.H., Syvitski, J.P.M., Wiberg, P.L. (Eds.),Continental-Margin Sedimentation: From Sediment Transport to Sequence Stra-tigraphy. International Association of Sedimentologists, Special Publication 37,pp. 49–100.

Hinderer, M., 2001. Late quaternary denudation of the Alps, valley and lake fillings andmodern river loads. Geodin. Acta 14 (4), 231–263.

Hinderer, M., 2012. From gullies to mountain belts: a review of sediment budgets atvarious scales. Sediment. Geol. 280, 21–59.

Holbrook, J., Bhattacharya, J.P., 2012. Reappraisal of the sequence boundary in timeand space: case and considerations for an SU (subaerial unconformity) that isnot a sediment bypass surface, a time barrier, or an unconformity. Earth Sci.Rev. 113, 271–302.

Holbrook, J., Wanas, H., 2014. A fulcrum approach to assessing source-to-sink mass bal-ance using channel paleohydrologic parameters derivable from common fluvialdata sets with an example from the cretaceous of Egypt. J. Sediment. Res. 84 (5),349–372.

Holbrook, J., Scott, R.W., Oboh-Ikuenobe, F.E., 2006. Base-level buffers and buttresses: amodel for upstream versus downstream control on fluvial geometry and architecturewithin sequences. J. Sediment. Res. 76 (1), 162–174.

Ielpi, A., Ghinassi, M., 2014. Planform architecture, stratigraphic signature andmorphodynamics of an exhumed Jurassic meander plain (Scalby formation,Yorkshire, UK). Sedimentology 61 (7), 1923–1960.

Ielpi, A., Ghinassi, M., 2015. Planview style and palaeodrainage of Torridonian channelbelts: Applecross formation, Stoer Peninsula, Scotland. Sediment. Geol. 325 (2015),1–16.







































http://dx.doi.org/10.1029/2007JF000851






















































































































Irving, E., Wynne, P.J., Globerman, B.R., 1993. Cretaceous paleolatitudes and over-prints of north American craton. In: Caldwell, W.G.E., Kauffman, E.G. (Eds.), Evo-lution of the Western Interior Basin. Geological Association of Canada, SpecialPaper 39, pp. 91–96.

Jaeger, J.M., Koppes, M., 2015. The Role of the Cryosphere in Source-to-Sink Systems.Earth-Science Reviews.

Jenkyns, H.C., 1980. Cretaceous anoxic events: from continents to oceans. J. Geol. Soc. 137(2), 171–188.

Jervey, M.T., 1988. Quantitative geological modeling of siliciclastic rock sequencesand their seismic expression. In: Wilgus, C.K., Hastings, B.S., Kendall, C.G.S.C.,Posamentier, H.W., Ross, C.A., Van Wagoner, J.C. (Eds.), Sea-level changes: an inte-grated approach. SEPM spec. Publ. 42, pp. 47–69.

Johnsson, M.J., 1990a. Overlooked sedimentary particles from tropical weathering envi-ronments. Geology 18 (2), 107–110.

Johnsson, M.J., 1990b. Tectonic versus chemical-weathering controls on the compositionof fluvial sands in tropical environments. Sedimentology 37 (4), 713–726.

Johnsson, M.J., 1993. The system controlling the composition of clastic sediments. Geol.Soc. Am. Spec. Pap. 284, 1–20.

Johnston, S.T., 2008. The cordilleran ribbon continent of North America. Annu. Rev. EarthPlanet. Sci. 36, 495–530.

Jorry, S.J., Droxler, A.W., Mallarino, G., Dickens, G.R., Bentley, S.J., Beaufort, L.,Peterson, L., Opdyke, B.N., 2008. Bundled turbidite deposition in the central Pan-dora trough (Gulf of Papua) since last glacial maximum: linking sediment natureand accumulation to sea level fluctuations at millennial timescale. J. Geophys.Res. Earth Surf. 113 (F1).

Kauffman, E.G., 1977. Geological and biological overview: Western Interior Cretaceousbasin. Mt. Geol. 14 (3–4), 75–99.

Kuehl, S.A., Alexander, C.R., Blair, N.E., Harris, C.K., Marsaglia, K.M., Ogston, A.S., orpin, A.R.,Roerig, J.J., Bever, A.J., Bilderback, E.L., Carter, L., Cerovski-Darriau, C., Laurel, B.,Childress, L.B., Corbett, D.R., Hale, R.P., Leithold, E.L., Litchfield, N., Moriart, J.M.,Page, M.J., Pierce, L.E.R., Upton, P., Walsh, J.P., 2015. A source to sink perspective ofthe Waipaoa River margin. Earth-Science Reviews.

Krause, F.F., Collins, H.N., Nelson, D.A., Machemer, S.D., French, P.R., 1987. Multiscaleanatomy of a reservoir: geological characterization of Pembina-Cardium pool, west-central Alberta, Canada. AAPG Bull. 71 (10), 1233–1260.

Krause, F. F., Deutsch, K. B., Joiner, S. D., Barclay, J. E., Hall, R. L., & Hills, L. V., 1994. Creta-ceous Cardium Formation of theWestern Canada sedimentary basin. in: Mossop, G.D.and Shetson, I., (Compilers), Geological Atlas of the Western Canada SedimentaryBasin. Canadian Society of Petroleum Geologists and Alberta Research Council,pp.485–511.

Kukulski, R.B., Hubbard, S.M., Moslow, T.F., Raines, M.K., 2013. Basin-scale stratigraphicarchitecture of upstream fluvial deposits: Jurassic–cretaceous foredeep, Albertabasin, Canada. J. Sediment. Res. 83 (8), 704–722.

Larue, D.K., Hovadik, J., 2006. Connectivity of channelized reservoirs: a modellingapproach. Pet. Geosci. 12 (4), 291–308.

Lawton, T.F., 2014. Small grains, big rivers, continental concepts. Geology 42 (7), 639–640.Lawton, T.F., Boyer, S.E., Schmitt, J.G., 1994. Influence of inherited taper on structural var-

iability and conglomerate distribution, Cordilleran fold and thrust belt, westernUnited States. Geology 22, 339–342.

Lawton, T.F., Pollock, S.L., Robinson, R.A.J., 2003. Integrating sandstone petrology andnonmarine sequence stratigraphy: application to the Late Cretaceous fluvial systemsof southwestern Utah, USA. J. Sediment. Res. 73 (3), 398–406.

Lawton, T.F., Bradford, I.A., Vega, F.J., Gehrels, G.E., Amato, J.M., 2009. Provenance of UpperCretaceous–Paleogene sandstones in the foreland basin system of the Sierra MadreOriental, northeastern Mexico, and its bearing on fluvial dispersal systems of theMexican laramide province. Geol. Soc. Am. Bull. 121 (5–6), 820–836.

Lawton, T.F., Eaton, J.G., Godfrey, K.N., Schellenbach, W.L., 2014a. Compositional, paleon-tological and detrital-zircon data from Cretaceous strata of the Henry Mountainsbasin and implications for connections with dispersal systems of Wahweap andKaiparowits formations in southern Utah, U.S.A. In: MacLean, J.S., Biek, R.F.,Huntoon, J.E. (Eds.), Geology of Utah's Far South: Salt Lake City, Utah. Utah GeologicalAssociation Publication 43, pp. 373–395 (compact disc).

Lawton, T.F., Schellenbach, W.L., Nugent, A.E., 2014b. Late Cretaceous fluvial-megafan andaxial-river systems in the southern Cordilleran foreland basin: drip tank member ofstraight cliffs formation and adjacent strata, southern Utah, USA. J. Sediment. Res.84, 407–434.

Lawton, Timothy F., Buller, Cody D., Parr, Todd R., 2015. Provenance of a Permian erg onthe western margin of Pangea: Depositional system of the Kungurian (lateLeonardian) Castle Valley and White Rim sandstones and subjacent Cutler Group,Paradox Basin, Utah, USA. Geosphere GES01174-1.

Leckie, D.A., 1986. Rates, controls, and sand-body geometries of transgressive-regressivecycles: Cretaceous moosebar and gates formations, British Columbia. AAPG Bull. 70(5), 516–535.

Leckie, D.A., Singh, C., 1991. Estuarine deposits of the Albian paddy member (Peace riverformation) and lowermost Shaftesbury formation, Alberta, Canada. J. Sediment. Res.61 (5), 825–849.

Leclair, S.F., Bridge, J.S., 2001. Quantitative interpretation of sedimentary structuresformed by river dunes. J. Sediment. Res. 71 (5), 713–716.

Lee, J.K.W., Williams, I.S., Ellis, D.J., 1997. Pb, U and Th diffusion in natural zircon. Nature390, 159–162.

Lee, K., McMechan, G., Gani, M.R., Bhattacharya, J.P., Zeng, X., Howell Jr., C.D., 2007. 3-D ar-chitecture and sequence stratigraphic evolution of a forced regressive top-truncatedmixed-influenced delta, Cretaceous Wall Creek Sandstone, Wyoming. J. Sediment.Res. 77, 303–323.

Leithold, E.L., Blair, N.E., Wegmann, K.W., 2015. Source to sink sedimentary systems andthe global C-cycle: A river runs through it. Earth-Science Reviews.

Li, W., Bhattacharya, J.P., Campbell, C., 2010. Temporal evolution of fluvial style withinacompound incised valley, The Ferron Notom Delta, Henry Mountains Region, Utah.J. Sediment. Res. 80, 529–549.

Li, Y., Bhattacharya, J.P., 2013. Falling stage compound valley systems, Ferron NotomDelta. J. Sediment. Res. 83, 206–225.

Li, Z., Bhattacharya, J.P., Schieber, J., 2015. Evaluating along-strike variation using thin-bedded facies analysis, Upper Cretaceous Ferron Notom Delta, Utah. Sedimentologyhttp://dx.doi.org/10.1111/sed.12219.

Liu, J.P., Xue, Z., Ross, K., Wang, H.J., Yang, Z.S., Li, A.C., Gao, S., 2009. Fate of sedimentsdelivered to the sea by Asian large rivers: long-distance transport and formation ofremote alongshore clinothems. The Sedimentary Record 7, pp. 4–9.

Long, D.G.F., 2004. Precambrian Rivers. The Precambrian Earth: Tempos and Events,pp. 660–663.

Long, D.G.F., 2011. Architecture and depositional style of fluvial systems before landplants: a comparison of Precambrian, early Paleozoic, and modern river deposits.In: Davidson, S.K., Leleu, S., North, C. (Eds.), From River to Rock Record. SEPM Spec.Publ 97, pp. 37–61.

Longman, M., Kneller, S., Meyer, T., Chapin, M., 2015. Pinedale field: Case study of a gianttight gas sandstone reservoir. AAPG Mem. 107 (531 pp.).

Lupe, R., Silberling, N.J., 1985. Genetic relationship between lower Mesozoic continentalstrata of the Colorado plateau and marine strata of the western great basin: signifi-cance for accretionary history of Cordilleran lithotectonic terranes. In: Howell, D.G.(Ed.), Tectonostratigraphic Terranes of the Circum-Pacific Region. Circum-PacificCouncil for Energy and Mineral Resources, Series 1, pp. 263–271.

Mack, G.H., 1984. Exceptions to the relationship between plate-tectonics and sandstonecomposition. J. Sediment. Petrol. 54, 212–220.

Macquaker, J.H.S., Bentley, S.J., Bohacs, K.M., 2010. Wave-enhanced sediment-gravityflows and mud dispersal across continental shelves: reappraising sediment transportprocesses operating in ancient mudstone successions. Geology 38, 947–950.

Magoon, L.B., Dow,W.G., 1994. The petroleum system. In: Magoon, L.B., Dow, W.G. (Eds.),The Petroleum System—From Source to Trap. AAPG Memoir 60, pp. 3–24.

Malkowski, M.A., Hampton, B.A., 2014. Sedimentology, U–Pb detrital geochronology, andHf isotopic analyses from Mississippian–Permian stratigraphy of the mysticsubterrane, Farewell terrane, Alaska. Lithosphere http://dx.doi.org/10.1130/L365.1.

Mallarino, G., Beaubouef, R.T., Droxler, A.W., Abreu, V., Labeyrie, L., 2006. Sea level influ-ence on the nature and timing of a minibasin sedimentary fill (northwestern slopeof the gulf of Mexico). AAPG Bull. 90 (7), 1089–1119.

Marsaglia, K.M., Ingersoll, R.V., 1992. Compositional trends in arc-related, deep-marinesand and sandstone: a reassessment of magmatic-arc provenance. Geol. Soc. Am.Bull. 104 (12), 1637–1649.

Marsaglia, K.M., DeVaughn, A.M., James, D.E., Marden, M., 2010. Provenance of fluvialterrace sediments within the Waipaoa sedimentary system and their importance toNew Zealand source-to-sink studies. Mar. Geol. 270 (1), 84–93.

Martinsen, R.S., 2003a. Depositional remnants, part 1: common components of thestratigraphic record with important implications for hydrocarbon exploration andproduction. AAPG Bull. 87 (12), 1869–1882.

Martinsen, R.S., 2003b. Depositional remnants, part 2: examples from theWestern Interi-or Cretaceous basin of North America. AAPG Bull. 87 (12), 1883–1909.

Martinsen, O.J., Ryseth, A.L.F., Helland-Hansen, W., Flesche, H., Torkildsen, G., Idil, S., 1999.Stratigraphic base level and fluvial architecture: Ericson sandstone (Campanian),rock springs uplift, SW Wyoming, USA. Sedimentology 46 (2), 235–263.

Matthai, H.F., 1990. Floods. Surface Water Hydrology. Geological Society of America,Boulder, Colorado, pp. 97–120.

McCarthy, P.J., 2002. Micromorphology and development of interfluve paleosols: a casestudy from the Cenomanian dunvegan formation, NE British Columbia, Canada.Bull. Can. Petrol. Geol. 50, 158–177.

McCarthy, P.J., Plint, A.G., 1999. Floodplain palaeosols of the Cenomanian Dunvegan for-mation, Alberta and British Columbia, Canada: micromorphology, pedogenic process-es and palaeoenvironmental implications. In: Marriott, S., Alexander, J., Hey, R. (Eds.),Floodplains: Interdisciplinary Approaches. Geological Society, London, Special Publi-cation 163, pp. 289–310.

McDonnell, A., Loucks, R.G., Galloway, W.E., 2008. Paleocene to Eocene deep-water slopecanyons, western gulf of Mexico: further insights for the provenance of deep-wateroffshore Wilcox group plays. AAPG Bull. 92 (9), 1169–1189.

Miall, A.D., 1997. The Geology of Stratigraphic Sequences. Springer.Miall, A.D., 2014. Fluvial Depositional Systems. Springer (316 pp.).Miall, A.D., 2015. Updating uniformitarianism: stratigraphy as just a set of ‘frozen

accidents’. Geol. Soc. Lond., Spec. Publ. 404 (1), 11–36.Miller, E.L., Soloviev, A.V., Prokopiev, A.V., Toro, J., Harris, D., Kuzmichev, A.B., Gehrels, G.E.,

2013. Triassic river systems and the paleo-Pacific margin of northwestern Pangea.Gondwana Res. 23, 1631–1645.

Milliman, J.D., Syvitski, J.P., 1992. Geomorphic/tectonic control of sediment discharge tothe ocean: the importance of small mountainous rivers. J. Geol. 525–544.

Moecher, D.P., Samson, S.D., 2006. Differential zircon fertility of source terranes andnatural bias in the detrital zircon record: implications for sedimentary provenanceanalysis. Earth Planet. Sci. Lett. 247 (3), 252–266.

Mulder, T., Syvitski, J.P., 1995. Turbidity currents generated at river mouths duringexceptional discharges to the world oceans. J. Geol. 285-299.

Mulder, T., Syvitski, J.P.M., Migeon, S., Faugères, J.-C., Savoye, B., 2003. Marine hyperpycnalflows: initiation, behavior and related deposits. A review. Mar. Pet. Geol. 20, 861–882.

Muller, D.S., Wirnkar, F.T., 2004. Stratigraphic entrapment of hydrocarbons in the UpperCretaceous Lewis shale and lower fox hills sandstone, Eastern Green River Basin,Wyoming. Houston Geological Society Bulletin 46, pp. 25–27.

Musial, G., Reynaud, J.Y., Gingras, M.K., Féniès, H., Labourdette, R., Parize, O., 2012. Subsur-face and outcrop characterization of large tidally influenced point bars of thecretaceous McMurray formation (Alberta, Canada). Sediment. Geol. 279, 156–172.



















































































http://dx.doi.org/10.1111/sed.12219
























http://dx.doi.org/10.1130/L365.1




















































Neal, J., Abreu, V., 2009. Sequence stratigraphy hierarchy and the accommodation succes-sion method. Geology 37 (9), 779–782.

Nystuen, J.P., 1998. History and development of sequence stratigraphy. In: Gradstein, F.M.,Sandvik, K.O., Milton, N.J. (Eds.), Sequence stratigraphy – Concepts and ApplicationNorwegian Petrol. Soc. Spec. Publ. 8, pp. 31–116.

O'Connell, S. C., 1994. Geological history of the Peace River arch. in: Mossop, G.D., andShetsen, I., (comp.), Geological Atlas of theWestern Canada Sedimentary Basin, Cana-dian Society of Petroleum Geologists and Alberta Research Council, pp. 431–438.

Olariu, M.I., Aiken, C.L.V., Bhattacharya, J.P., Xu, X., 2011. Interpretation of channelizedarchitecture using three-dimensional photo real models, Pennsylvanian deep-waterdeposits at big rock quarry, Arkansas. Mar. Pet. Geol. 28 (6), 1157–1170.

Oliver, W.B., 1970. Depositional systems in the woodbine formation (Upper Cretaceous),northeast Texas. Gulf Coast Association of Geological Societies Transactions 20(262–262).

Owen, A., Nichols, G.J., Hartley, A.J., Weissmann, G.S., Scuderi, L.A., 2015. Quantification ofa distributive fluvial system: the salt wash DFS of the Morrison formation, SW USA.J. Sediment. Res. 85 (5), 544–561.

Painter, C.S., Carrapa, B., DeCelles, P.G., Gehrels, G.E., Thomson, S.N., 2014. Exhumation ofthe north American cordillera revealed by multi-dating of upper Jurassic–upper cre-taceous foreland basin deposits. Geol. Soc. Am. Bull. http://dx.doi.org/10.1130/B30999.1.

Parsons, J.D., Bush, J.W.M., Syvitski, J.P.M., 2001. Hyperpycnal plume formation from riv-erine outflows with small sediment concentrations. Sedimentology 48, 465–478.

Petter, A.L., Steel, R.J., Mohrig, D., Kim, W., Carvajal, C., 2013. Estimation of the paleoflux ofterrestrial-derived solids across ancient basin margins using the stratigraphic record.Geol. Soc. Am. Bull. 125 (3–4), 578–593.

Pettijohn, F.J., Potter, P.E., Siever, R., 1987. Sand and Sandstone. Springer-Verlag, New York(553 pp.).

Plint, A.G., 1988. Sharp-based shoreface sequences and “offshore bars” in the Cardium for-mation of Alberta: their relationship to relative changes in sea level. In: Wilgus, C.K.,Hastings, B.S., Kendall, C.G.S.C., Posamentier, H.W., Ross, C.A., Van Wagoner, J.C.(Eds.), Sea-level changes: an integrated approach SEPM Special Publication 42,pp. 357–370.

Plint, A.G., 2000. Sequence stratigraphy and paleogeography of a Cenomanian deltaiccomplex: the Dunvegan and lower Kaskapau formations in subsurface and outcrop,Alberta and British Columbia, Canada. Bull. Can. Petrol. Geol. 47, 43–79.

Plint, A.G., 2002. Paleovalley systems in the Upper Cretaceous Dunvegan formation,Alberta and British Columbia. Bull. Can. Petrol. Geol. 50, 277–296.

Plint, A.G., 2010. Wave-and storm-dominated shoreline and shallow-marine systems. In:Dalrymple, R.W., James, N.P. (Eds.), Facies Models, 4, Geological Association ofCanada. Geotext 6, pp. 167–201.

Plint, A.G., Wadsworth, J.A., 2003. Sedimentology and Palaeogeomorphology of FourLarge Valley Systems Incising Delta Plains, Western Canada Foreland Basin; Im-plications for mid-Cretaceous sea-level changes. Sedimentology 50, 1147–1186.

Plint, A.G., Wadsworth, J.A., 2006. Delta plain paleodrainage patterns reflect small-scalefault movement and subtle forebulge uplift: upper cretaceous dunvegan formation,Western Canada Foreland Basin. In: Dalrymple, R.W., Leckie, D.A., Tillman, R.W.(Eds.), Incised-Valley Systems in Time and Space. SEPM, Special Publication 85,pp. 219–237.

Plint, A.G., Walker, R.G., Bergman, K.M., 1986. Cardium formation 6. Stratigraphic frame-work of the cardium in subsurface. Bull. Can. Petrol. Geol. 34 (2), 213–225.

Plint, A.G., Walker, R.G., Duke, W.L., 1988. An outcrop to subsurface correlation of thecardium formation in Alberta. In: James, D.P., Leckie, D.A. (Eds.), Sequences, Stratigra-phy, Sedimentology: Surface and Subsurface. Canadian Society of Petroleum Geolo-gists Memoir 15, pp. 167–183.

Plint, A.G., Tyagi, A., Hay, M.J., Varban, B.L., Zhang, H., Roca, X., 2009. Clinoforms,paleobathymetry, and mud dispersal across the western Canada cretaceous forelandbasin: evidence from the Cenomanian dunvegan formation and contiguous strata.J. Sediment. Res. 79 (3), 144–161.

Posamentier, H.W., Allen, G.P., 1993. Variability of the sequence stratigraphic model:effects of local basin factors. Sediment. Geol. 86 (1), 91–109.

Posamentier, H.W., Walker, R.G., 2006. Deep water turbidites and turbiditic fans. In:Posamentier, H.W., Walker, R.G. (Eds.), Facies Models Revisited. SEPM Special Publi-cation 84, pp. 339–527.

Posamentier, H.W., Jervey, M.T., Vail, P.R., 1988. Eustatic controls on clastic deposition I –conceptual framework. In: Wilgus, C.K., Hastings, B.S., Kendall, C.G.S.C., Posamentier,H.W., Ross, C.A., Van Wagoner, J.C. (Eds.), Sea Level Changes – An IntegratedApproach. SEPM Special Publication 42, pp. 110–124.

Pranter, M.J., Cole, R.D., Panjaitan, H., Sommer, N.K., 2009. Sandstone-body dimensions ina lower coastal-plain depositional setting: lower Williams fork formation, coal can-yon, piceance basin, Colorado. AAPG Bull. 93 (10), 1379–1401.

Qayyum, F., De Groot, P., Nanne Hemstra, N., Octavian Catuneanu, O., 2015. 4Dwheeler diagrams: concept and applications. Geol. Soc. Lond., Spec. Publ. 404,223–232.

Rainbird, R.H., Hearnan, L.M., Young, G., 1992. Sampling laurentia: detrital zircon geochro-nology offers evidence for an extensive Neoproterozoic river system originating fromthe Grenville orogen. Geology 20 (4), 351–354.

Raines, M.K., Hubbard, S.M., Kukulski, R.B., Leier, A.L., Gehrels, G.E., 2013. Sediment dis-persal in an evolving foreland: detrital zircon geochronology from upper Jurassicand lowermost cretaceous strata, Alberta basin, Canada. Geol. Soc. Am. Bull. 125(5–6), 741–755.

Reijenstein, H.M., Posamentier, H.W., Bhattacharya, J.P., 2011. Seismic geomorphologyand high-resolution seismic stratigraphy of inner-shelf fluvial, estuarine, deltaic,and marine sequences, gulf of Thailand. AAPG Bull. 95 (11), 1959–1990.

Reinson, G. E., Warters, W. J., Cox, J., & Price, P. R., 1994. Cretaceous Viking formation of thewestern Canada sedimentary basin. in: Mossop, G.D. and Shetsen, I., (Comp.),

Geological Atlas of theWestern Canada sedimentary basin: Calgary, Canadian Societyof Petroleum Geologists and Alberta Research Council, pp.353–364.

Reynolds, A.D., 1999. Dimensions of paralic sandstone bodies. AAPG Bull. 83 (2),211–229.

Ricci-Lucchi, F.R., 1995. Contessa and associated megaturbidites: long distance(120 × 25 km) correlation of individual beds in a Miocene foredeep. In:Pickering, K.T., Hiscott, R.N., Kenyon, N.H., Ricci Lucchi, F., Smith, R.D.A. (Eds.),Atlas of Deep Water Environments. Springer, Netherlands, pp. 300–302.

Riggs, N.R., Lehman, T.M., Gehrels, G.E., Dickinson, W.R., 1996. Detrital zircon linkbetween headwaters and terminus of the Upper Triassic Chinle-Dockum paleoriversystem. Science 273 (5271), 97–100.

Rijn, L.C.V., 1984. Sediment transport, part II: suspended load transport. J. Hydraul. Eng.110 (11), 1613–1641.

Rine, J.M., Ginsburg, R.N., 1985. Depositional facies of a mud shoreface in Suriname, SouthAmerica: a mud analogue to sandy, shallow-marine deposits. J. Sediment. Petrol. 55,633–652.

Robinson, J.W., McCabe, P.J., 1997. Sandstone-body and shale-body dimensions in abraided fluvial system: salt wash sandstone member (Morrison formation), Garfieldcounty, Utah. AAPG Bull. 81 (8), 1267–1291.

Romans, Brian W., Castelltort, Sebastien, Covault, Jacob A., Fildani, Andrea, Walsh, J.P.,2015. Environmental signal propagation in sedimentary systems across timescales.Earth-Science Reviews.

Rubin, D.M., McCulloch, D.S., 1980. Single and superimposed bedforms: a synthesis of SanFrancisco Bay and flume observations. Sediment. Geol. 26, 207–231.

Sadler, P.M., 1981. Sediment accumulation rates and the completeness of stratigraphicsections. J. Geol. 569–584.

Schieber, J., Southard, J., Thaisen, K., 2007. Accretion of mudstone beds from migratingfloccule ripples. Science 318 (5857), 1760–1763.

Schlanger, S.O., Jenkyns, H.C., 1976. Cretaceous oceanic anoxic events: causes and conse-quences. Geol. Mijnb. 55, 179–194.

Singer, B.S., Meyers, S.R., Sagemen, B.B., Jicha, B.R., Condon, D., 2015. Improving Creta-ceous time scale uncertainties via multi-collector 40Ar/39Ar dating. GSA AnnualMeeting Abstracts.

Slingerland, R., Driscoll, N.W., Milliman, J.D., Miller, S.R., Johnstone, E.A., 2008. Anatomyand growth of a Holocene clinothem in the Gulf of Papua. J. Geophys. Res. EarthSurf. 113 (F1).

Smith, D. G., 1994. Paleogeographic evolution of the Western Canada foreland basin. inMossop, G.D., and Shetsen, I. (comps.), Geological Atlas of the Western Canadasedimentary basin: Canadian Society of Petroleum Geologists and Alberta ResearchCouncil, pp. 277–296.

Smith, D.G., Hubbard, S., Leckie, D., Fustic, M., 2009. Counter point bars in modernmeandering rivers: recognition of morphology, lithofacies and reservoir significance,examples from Peace river, AB, Canada. Sedimentology 56, 1655–1669.

Sømme, T.O., Jackson, C.A.L., 2013. Source-to-sink analysis of ancient sedimentary sys-tems using a subsurface case study from the møre-trøndelag area of southernNorway: part 2– sediment dispersal and forcing mechanisms. Basin Res. 25 (5),512–531.

Sømme, T.O., Helland-Hansen, W., Martinsen, O.J., Thurmond, J.B., 2009. Relationshipsbetween morphological and sedimentological parameters in source-to-sink systems:a basis for predicting semi-quantitative characteristics in subsurface systems. BasinRes. 21 (4), 361–387.

Sømme, T.O., Piper, D.J., Deptuck, M.E., Helland-Hansen, W., 2011. Linking onshore–offshore sediment dispersal in the golo source-to-sink system (Corsica, France)during the Late Quaternary. J. Sediment. Res. 81 (2), 118–137.

Sømme, T.O., Jackson, C.A.L., Vaksdal, M., 2013. Source-to-sink analysis of ancient sedi-mentary systems using a subsurface case study from the møre-trøndelag area ofsouthern Norway: part 1–depositional setting and fan evolution. Basin Res. 25 (5),489–511.

Soyinka, O.A., Slatt, R.M., 2008. Identification and micro-stratigraphy of hyperpycnitesand turbidites in cretaceous Lewis shale, Wyoming. Sedimentology 55 (5), 117–1133.

Spencer, C.J., Prave, A.R., Cawood, P.A., Roberts, M.W., 2014. Detrital zircon geochronologyof the Grenville/Llano foreland and basal Sauk Sequence in west Texas, USA. Geol.Soc. Am. Bull. 126 (7–8), 1117–1128. http://dx.doi.org/10.1130/B30884.1.

Stelck, C. R., 1975. Basement control of cretaceous sand sequences in western Canada. In:Caldwell, W.G.E. (Ed.), The cretaceous system in the western interior of NorthAmerica. Geological Association of Canada, Special Paper, 13, 427–440.

Stott, D.F., 1982. Lower Cretaceous Fort St. John Group and Upper Cretaceous DunveganFormation of the Foothills and Plains of Alberta, British Columbia, District ofMackenzie and Yukon Territory. Geological Survey of Canada Bulletin 328 (124 pp.).

Strong, N., Paola, C., 2008. Valleys that never were: time surfaces versus stratigraphicsurfaces. J. Sediment. Res. 78 (8), 579–593.

Suter, J.R., 2006. In: Posamentier, H.W., Walker, R.G. (Eds.), Facies models revisited.Special Publication-SEPM 84, pp. 337–395.

Syvitski, J.P., Milliman, J.D., 2007. Geology, geography, and humans battle for dominanceover the delivery of fluvial sediment to the coastal ocean. J. Geol. 115 (1), 1–19.

Szwarc, T.S., Johnson, C.L., Stright, L.E., McFarlane, C.M., 2015. Interactions between axialand transverse drainage systems in the Late Cretaceous Cordilleran foreland basin:evidence from detrital zircons in the straight cliffs formation, southern Utah, USA.Geol. Soc. Am. Bull. 127, 372–392.

Thomas, R.G., Smith, D.G., Wood, J.M., Visser, J., Calverley-Range, E.A., Koster, E.H., 1987.Inclined heterolithic stratification—terminology, description, interpretation andsignificance. Sediment. Geol. 53 (1), 123–179.

Tye, R.S., 2004. Geomorphology: an approach to determining subsurface reservoir dimen-sions. AAPG Bull. 88 (8), 1123–1147.

Tye, R.S., Bhattacharya, J.P., Lorsong, J.A., Sindelar, S.T., Knock, D.G., Puls, D.D., Levinson,R.A., 1999. Geology and stratigraphy of fluvio-deltaic deposits in the Ivishak















http://dx.doi.org/10.1130/B30999.1

http://dx.doi.org/10.1130/B30999.1





















































































































http://dx.doi.org/10.1130/B30884.1




















formation: applications for development of Prudhoe Bay Field, Alaska. AAPG Bull. 83,1588–1623.

Ullah, M., Bhattacharya, J.P., Dupré, W.R., 2015. Confluence scours versus incised valleys:examples from the Cretaceous Ferron Notom Delta, SE Utah. J. Sediment. Res. 5 (5),445–458.

Vail, P.R., 1987. Seismic stratigraphy interpretation using sequence stratigraphy: part 1:seismic stratigraphy interpretation procedure. In: Bally, A.W. (Ed.)Atlas of SeismicStratigraphy AAPG Studies in Geology 27, pp. 1–10.

Vail, P.R., Mitchum Jr., R.M., Thompson III, S., 1977. Seismic stratigraphy and global chang-es of sea level, part 4: Global cycles of relative changes of sea level. In: Payton, C.E.(Ed.), Seismic Stratigraphy - Applications to Hydrocarbon Exploration. AmericanAssociation of Petroleum Geologists Memoir 26, pp. 83–97.

Vakarelov, B.K., Bhattacharya, J.P., 2009. Local tectonic control on parasequence architec-ture: second frontier sandstone, powder river basin, Wyoming. AAPG Bull. 93 (3),295–327.

van der Beek, P., Robert, W., Mugnier, J.-L., Bernet, M., Huyghe, P., Labrin, E., 2006. LateMiocene-recent exhumation of the central Himalaya and recycling in the forelandbasin assessed by apatite fission-track thermochronology of Siwalik sediments,Nepal. Basin Res. 18, 413–434.

Van Wagoner, J.C., Bertram, G.T. (Eds.), 1995. Sequence stratigraphy of foreland basin de-posits: outcrop and subsurface examples from the cretaceous of North America.AAPG Memoir 64 (487 pp.).

Van Wagoner, J.C., Mitchum, R.M., Campion, K.M., Rahmanian, V.D., 1990. Siliciclastic se-quence stratigraphy in well logs, cores, and outcrops: concepts for high-resolutioncorrelation of time and facies. AAPG methods in exploration series 7 (55 pp.).

Vermeesch, P., 2004. How many grains are needed for a provenance study? Earth planet.Sci. Lett. 224, 351–441.

Walker, R.G., 1992. Turbidites and submarine fans. In: Walker, R.G., James, N.P. (Eds.), Fa-cies models response to sea-level changes. Geological Association of Canada,pp. 239–263.

Walker, R.G., Eyles, C.H., 1988. Geometry and facies of stacked shallow-marine sandierupward sequences dissected by erosion surface, Cardium formation, Willesdengreen, Alberta. AAPG Bull. 72 (12), 1469–1494.

Walsh, J.P., Nittrouer, C.A., 2003. Contrasting styles of off-shelf sediment accumulation innew guinea. Mar. Geol. 196, 105–125.

Walsh, J.P., Nittrouer, C.A., 2009. Understanding fine-grained river-sediment dispersal oncontinental margins. Mar. Geol. 263, 34–45.

Walsh, J.P., Corbett, D.R., Ogston, A.S., Nittrouer, C., Kuehl, S.A., Allison, M.A., Goodbred Jr.,S.L., 2015. Shelf and slope sedimentation associated with large deltaic systems. In:

Bianchi, T.S, Allison, M.A, Cai, W (Eds.), Biogeochemical Dynamics at Major River-Coastal Interfaces: Linkages with Global Change. Cambridge University Press, NewYork, NY, p. 704.

Warne, A.G., Meade, R.H., White, W.A., Guevara, E.H., Gibeaut, J., Smyth, R.C., Aslan, A.,Tremblay, T., 2002. Regional controls on geomorphology, hydrology, and ecosystemintegrity in the Orinoco Delta, Venezuela. Geomorphology 44, 273–307.

Weissmann, G.S., Hartley, A.J., Nichols, G.J., Scuderi, L.A., Olson, M., Buehler, H., Banteah, R.,2010. Fluvial form in modern continental sedimentary basins: distributive fluvialsystems. Geology 38, 39–42.

Wernicke, B., 2011. The California river and its role in carving the Grand Canyon. Geol.Soc. Am. Bull. 123, 1288–1316.

Wescott, W.A., 1993. Geomorphic thresholds and complex response of fluvial systems–some implications for sequence stratigraphy. AAPG Bull. 77 (7), 1208–1218.

Williams, G.P., 1988. Paleofluvial estimates from dimensions of former channels andmeanders. In: Baker, V.R., Kochel, R.C., Patton, P.C. (Eds.), Flood Geomorphology.John Wiley & Sons, New York, pp. 321–334.

Williams, G.D., Stelck, C.R., 1975. Speculations on the Cretaceous paleogeography ofNorth America. In: Caldwell, W.G.E. (Ed.), The Cretaceous System in the WesternInterior of North America. The Geological Association of Canada Special PaperNo.13, pp. 1–20.

Winker, C.D., 1982. Cenozoic shelf margins, northwestern gulf of Mexico. Gulf CoastAssociation of Geological Societies Transactions 32, pp. 427–448.

Wright, S., Parker, G., 2004. Flow resistance and suspended load in sand-bed rivers:simplified stratification model. J. Hydraul. Eng. 130 (8), 796–805.

Wright, G. N., McMechan, M. E., & Potter, D. E. G., 1994. Structure and architecture of theWestern Canada sedimentary basin. in: Mossop, G.D. and Shetsen, I. (Comps.)Geological atlas of the Western Canada sedimentary basin, pp. 25–40.

Wu, C., Bhattacharya, J.P., Ullah, M., 2015. Paleohydrology and 3D facies architecture ofancient point bars, Ferron sandstone, Notom Delta, South-central, Utah. J. Sediment.Res. 85, 399–418.

Zaleha, M.J., 2013. Paleochannel hydraulics, geometries, and associated alluvial architec-ture of Early Cretaceous rivers, Sevier foreland basin, Wyoming, USA. Cretac. Res.45, 321–341.

Zhu, Y., Bhattacharya, J.P., Li, W., Lapen, T.J., Jicha, B., Singer, B., 2012. Milankovitch-scale sequence stratigraphy and stepped forced regressions of the TuronianFerron Notom deltaic complex, south-central Utah, U.S.A. J. Sediment. Res. 82,723–746.







































































estimation of source area, river paleo-discharge, …...invited review estimation of source area,...

Documents