Reevaluating the process regime in the Sego Sandstone: Sedimentological and ichnological evidence for an underemphasised fluvial signature

Rocks of coastal to shallow‐marine origin are challenging to interpret owing to the complex interplay of various depositional processes. This study reevaluates the relative roles of fluvial, tidal and wave processes in the Upper Cretaceous Sego Sandstone (and subordinately in the underlying Buck Tongue) of the Book Cliffs, USA, a well‐studied ancient coastal to shallow‐marine succession. Detailed sedimentological and ichnological analyses were used to interpret a previously underemphasised riverine signature, consisting of centimetre‐ to decimetre‐thick alternations of sandstone and heterolithic beds inferred to represent flood–interflood periods of variable river discharge. Recognition of a widespread fluvial‐dominated signature across the studied units better agrees with other sedimentological and regional observations in the study area, such as high sandstone–mudstone ratios, largely unidirectional and seaward‐oriented palaeocurrents, and modelled weak tidal conditions in the basin. When considering all of the sedimentological, ichnological and stratigraphic observations together with its regional depositional context, the Sego Sandstone/Buck Tongue system is better explained using a mixed‐energy but fluvial‐dominated deltaic model. This highlights an historical over‐interpretation of tidal processes and subordinate wave processes in the Sego Sandstone and likely in similar units. The widely used approach that emphasises only certain sedimentary features in discerning the process regime from analysis of rocks of inferred coastal to shallow‐marine origin is unrefined and may therefore underrepresent the actual complexity of these systems.


| INTRODUCTION
Since the 1970s, it has been suggested that modern coastal to shallow-marine systems are controlled by the relative roles of fluvial, tidal and wave processes (Boyd et al., 1992;Galloway, 1975;Orton & Reading, 1993).In a similar way, ancient analogue systems reconstructed from the geological record have also been typically assigned to one or more of the above-mentioned depositional processes (Legler et al., 2013(Legler et al., , 2014;;Rossi et al., 2017;Rossi & Steel, 2016;van Cappelle et al., 2016).The commonly used approach for interpreting coastal to shallow-marine rocks and their process regimes largely relies on 'process indicators', which are predominantly a subset of sedimentary structures considered diagnostic of a particular process.For example, hummocky cross-stratification and symmetric ripples are associated with wave processes, whereas mudstone drapes and cyclic patterns in cross-stratification are usually assigned to tidal processes (McIlroy et al., 2005;Onyenanu et al., 2018;Rossi & Steel, 2016;Vakarelov et al., 2012;van Cappelle et al., 2016).These are commonly used in conjunction with trace-fossil assemblages to infer environmental conditions associated with process regimes (Weleschuk & Dashtgard, 2019) and, in some cases, coupled with other observations (e.g. the origin of erosional scours; Willis & Gabel, 2003).The general concept of process indicators has also been used more recently to create semiquantitative or quantitative methods to assess the relative roles of depositional processes, by quantifying the distribution of sedimentary structures associated with each process along core or outcrop logs (Ainsworth et al., 2011;Rossi et al., 2017).In the revised ternary classification of Ainsworth et al. (2011), which can be used both qualitatively and quantitatively for modern and ancient systems, the primary, secondary and tertiary inferred processes are termed, respectively, dominated, influenced and affected (e.g.tide dominated, wave influenced, fluvial affected = Twf).
Although the ternary process diagram forms a useful framework, interpreting coastal to shallow-marine rocks is challenging, due to the inherent uncertainties of the geological record, notably concerning preservation potential of sedimentary structures coupled with variability and interactions of multiple depositional processes within these environments.Significant doubts remain regarding the scientific strength and merit of both the qualitative and quantitative approaches aimed at assessing depositional processes from the sedimentary record based principally on sedimentary structures.In fact, most sedimentary structures do not unequivocally record the action of a single process in a particular environment (Ainsworth et al., 2012;Visser, 1980), and there is no evidence of a one-to-one correspondence between either the strength or frequency of depositional processes and the preservation potential of their related sedimentary products (Ghinassi et al., 2018;Gugliotta et al., 2020).In addition, as it will be shown in this paper, stratigraphic architecture, sandstone-mudstone ratios, thicknesses, accretionary surfaces, interbedding styles and many other sedimentological features are often not considered, or else they receive much less emphasis when the process regime is assessed.Recent studies have shown that multiple explanations exist for sedimentary structures that were, in the past, considered diagnostic of a specific process.For example, cyclic patterns in cross-stratification characterised by fine-grained drapes on sand laminae were long considered to be indicative of tidal processes (Legler et al., 2013(Legler et al., , 2014;;van Cappelle et al., 2016;Visser, 1980).However, these also have been shown to be associated with a wider range of environments and processes, such as superimposed bedform dynamics, wind action and modulation of riverine currents (Ainsworth et al., 2012;Gugliotta et al., 2016a;Martinius & Gowland, 2011;Reesink & Bridge, 2011).Additionally, hummocky and swaley cross-stratification are also alternatively interpreted to potentially record the combination of oscillatory and unidirectional flows related to several processes, and not exclusively to wave storms (Collins et al., 2017;Jelby et al., 2020;Quin, 2011).Recent work on modern and ancient fluvialto-marine transition zones (FMTZs) has shown that deltaic and estuarine channels, as well as their adjacent areas, are characterised by complex dynamics between multiple contemporaneous processes and consequently, their resulting sedimentary structures cannot easily be tagged to be the result of a single process (Collins et al., 2020;Gugliotta et al., 2017Gugliotta et al., , 2019Gugliotta et al., , 2020;;Gugliotta et al., 2016a;2016b;Hendershot et al., 2016;Jablonski & Dalrymple, 2016;La Croix & Dashtgard, 2014, 2015;Martinius et al., 2015;Martinius & Gowland, 2011).
When studying tidal systems, these recent developments have necessitated a critical review of previous research.Several geological units, such as the Middle Jurassic Lajas Formation (Argentina), the Lower Jurassic Tilje Formation (offshore Norway) and the Lower Cretaceous McMurray Formation (Canada), are among the best-known interpreted tide-dominated systems (Martinius et al., 2001;McIlroy et al., 2005;Smith, 1988).All these units have since been reinterpreted as mixedenergy systems, wherein tidal processes were reconsidered as only one component of a more complex process regime, and in some cases deemed subordinate to other prevailing processes (Gugliotta et al., 2015;Gugliotta et al., 2016a;2016b;Ichaso et al., 2016;Ichaso & Dalrymple, 2009;Kurcinka et al., 2018;Martinius et al., 2015;Rossi & Steel, 2016).For ancient tidal systems, a key unit is the Sego Sandstone from the Book Cliffs in the USA (Figure 1A).The Sego is a well-documented example interpreted as an ancient tide-dominated delta and has been studied in detail by Willis andGabel (2001, 2003).Based on these studies on the Sego, a review paper on tide-influenced deltaic systems was published by Willis (2005).Studies interpreting similar systems became increasingly popular for rock units of different ages globally (Legler et al., 2013;McIlroy et al., 2005;Tänavsuu-Milkeviciene & Plink-Björklund, 2009).Although several open debates regarding the Sego remain-for example the general depositional environment, formational mechanisms of erosional surfaces, relationship with the Buck Tongue-there has been a wide and consistent agreement regarding the fundamental role of tidal process (alone or in combination with the wave process) during the accumulation of these deposits (Burton et al., 2016;Franczyk, 1989;Legler et al., 2014;Painter et al., 2013;van Cappelle et al., 2016;Willis, 2000;Willis & Gabel, 2001, 2003;York et al., 2011).
To date, this view of the Sego Sandstone as a strongly tidal system has never been directly challenged.In this paper, it is suggested that certain sedimentological features have been overlooked or at least underemphasised for decades and represent missing parts of the process 'puzzle', which calls into question the view of the Sego as a strongly tidal system.The aim of this paper is: (1) to report and interpret an overlooked interbedding pattern in the studied units; (2) to discuss the general interpretation of the studied units with implications for similar rocks; and (3) to suggest a revision of the approaches used to assess process regimes in coastal to shallow-marine rocks.

| GEOLOGICAL BACKGROUND AND HISTORICAL VIEWS AS A STRONGLY TIDAL SYSTEM
The Sego Sandstone (or Sego Member) is part of the Upper Cretaceous Mesaverde Group and outcrops along the Book Cliffs in Utah and Colorado, USA (Figure 1A).This unit is up to 80 m thick and largely consists of sandstone with subordinate mudstone that accumulated in a coastal to shallow-marine environment (van Cappelle et al., 2016;Willis & Gabel, 2001).It is subdivided into lower and upper parts, which are separated by the intervening Anchor Mine Tongue (part of the Mancos Shale) to the east (Figure 1B).The Sego Sandstone overlies the shelfal mudstone deposits of the Buck Tongue (also part of the Mancos Shale) and is overlain by coastal-plain deposits of the Neslen Formation and Mount Garfield Formation.The unit also transitions laterally to the fluvial deposits of the middle Castlegate Sandstone to the west (i.e.palaeolandward) and to the shelf deposits of the Mancos Shale to the east (i.e.palaeoseaward) (Legler et al., 2014;van Cappelle et al., 2016).The deposits of the Sego Sandstone accumulated during the middle to late Campanian and reflect progradation roughly eastwards within the Utah Bight on the western margin of the Western Interior Seaway (WIS; Figure 2).This area was part of the foreland basin of the North American Cordillera experiencing temperate climates with hot summers and cool winters (Sellwood & Valdes, 2006;van Cappelle et al., 2016).
The Sego Sandstone has been intensely studied sedimentologically for more than three decades and is widely regarded as a world-class example of an ancient tidal or mixed tidal and wave coastal to shallow-marine system (Table 1).The first detailed sedimentological study with a focus on the Sego Sandstone is thought to have been conducted by Franczyk (1989), who interpreted the palaeocoastline primarily as a barrier-island system.In that model, the sandstone-rich deposits of the Sego Sandstone were regarded as having accumulated in a tide-influenced, back-barrier environment that was located landward of the coeval coastline.Later, van Wagoner (1991a, 1991b) reinterpreted these deposits as tide-dominated deltas (although this view is sometimes incorrectly reported as estuarine, e.g.p. 438 of Steel et al., 2012) filling stacked incised valleys.van Wagoner (1991avan Wagoner ( , 1991b) identified regional unconformities at the base of the valleys that were explained by relative sea-level changes in a sequence stratigraphic context.The interpretation of incised-valley fills was successively supported by Willis (2000), although the deposits of the Sego Sandstone were interpreted as estuarine, and the regional unconformities attributed to isostatic rebound.
In the early 2000s, Willis andGabel (2001, 2003) proposed that the Sego Sandstone originated as an ancient, tide-dominated deltaic system with erosional surfaces at the base of the sandstone bodies explained by tidal scouring rather than incision of valleys.Comparisons regarding the tidal scours were made with modern tidedominated deltas such as the Fly, Ganges-Brahmaputra and Yangtze (Willis & Gabel, 2003).The interpretation as a deltaic system was further supported by Legler et al. ( 2014) and later by van Cappelle et al. ( 2016), although in both these studies the Sego Sandstone and underlying Buck Tongue were considered to reflect a mixed tide-influenced and wave-influenced deltaic system rather than simply tide dominated.In these studies, the emphasis for a large portion of the facies in the Sego Sandstone remained on tidal dominance, with subordinate fluvial influence, whereas other facies, predominantly in the Buck Tongue, were considered to be wave dominated.According to van Cappelle et al. (2016), the depositional environment of the Sego Sandstone/Buck Tongue system would have been at least in part comparable to modern mixed-energy to tide-dominated river deltas, such as the Fly, Mahakam, Mekong and Niger, as well as with the conceptual model for the FMTZ of tide-dominated deltas published by Dalrymple and Choi (2007).Similarities between the Sego Sandstone/ Buck Tongue system and the modern Mekong River delta were also reported by Legler et al. (2014).
Over the last decade or so, other detailed sedimentological studies on the Sego Sandstone have provided interpretations mostly in line with the previous literature, where depositional environments varied among deltas, estuaries and barrier systems, and with a strong emphasis on tidal or mixed tidal and wave processes (Burton et al., 2016;Burton & Wood, 2011;Painter et al., 2013;York et al., 2011).Recently, Rossi et al. (2017) applied their quantitative method to two 8 m and 17 m thick sections of the lower Sego Sandstone based on the sedimentological logs from Legler et al. (2014), concluding that these portions of the system varied between a tidedominated and wave-dominated classification, with varying degrees of influence by subordinate processes.The view of the Sego Sandstone as a strongly tidal estuarine or deltaic system remained widely accepted in other studies on the Book Cliffs and WIS (Aschoff & Steel, 2011a, 2011b;Li & Aschoff, 2022;McLaurin & Steel, 2000;Steel et al., 2012;Yoshida, 2000;Yoshida et al., 1996).Only in a recent regional study did van Cappelle et al. (2018) report for the first time the shoreline of the Sego to be fluvial dominated, tide influenced and wave influenced (see their figure 7); however, this appears in contrast with the literature mentioned above, especially in light of the detailed sedimentological studies conducted by the same research group a few years before (i.e.Legler et al., 2014;van Cappelle et al., 2016), in which tidal and wave processes were emphasised and modern analogues depicting strong tidal processes were chosen.Tidal modelling for the WIS shows that tides were rather minor in this epicontinental sea during the Late Cretaceous, with microtidal conditions and weak bed shear stresses in the Utah Bight, regardless of changes to large-scale palaeobathymetry (Dean et al., 2019;Ericksen & Slingerland, 1990).Nonetheless, Dean et al. (2019) suggested that local  confinements could be used to justify the amplification of tides in certain areas to explain the commonly reported tidal deposits in the Sego Sandstone, maintaining an emphasis on potentially strong tides despite the modelling results.

| DATA AND METHODS
The present study was conducted with field data collection in different areas of the Book Cliffs combining traditional sedimentological logging, facies analysis and examination of body fossils, trace fossils and bioturbation intensities (Bioturbation Index, BI).Sedimentological logs were compiled at selected outcrops in Utah (Tusher Canyon, Crescent Canyon and Thompson Canyon; Figure 1A), with additional observations from other localities (e.g.Sego Canyon, San Arroyo Canyon) and supplemented with the review of the existing literature.The main observations were made for the lower Sego Sandstone (the main target of this study) and to a lesser degree for the underlying Buck Tongue (Figures 1B and 3).Five facies associations (FAs) were identified and interpreted as different depositional elements of a deltaic system.No major differences in terms of general stratigraphy and depositional elements (e.g.channels, bars) are presented herein compared to other recent papers (Legler et al., 2014;van Cappelle et al., 2016); thus, the FAs are only briefly summarised below.Nevertheless, fundamentally different interpretations regarding the process regime are presented and the nomenclature was adapted following these reinterpretations (e.g.distributary channels instead of tidal channels; mouth bars instead of tidal bars).

| FACIES ASSOCIATIONS
The studied deposits are organised in a general coarseningupward trend represented by increases in grain size, sandstone-mudstone ratios and bed thicknesses, and show internal packages (i.e.parasequences) at the decametrescale with similar vertical trends (Figure 4).The five identified FAs consist of distributary-channel deposits (FA 1), mouth-bar deposits (FA 2), delta-front deposits (FA 3), reworked delta-front deposits (FA 4) and prodelta deposits (FA 5), and are described below.Facies Association 1 and FA 2 are part of the lower Sego Sandstone, whereas FA 3, FA 4 and FA 5 are part of the Buck Tongue.
T A B L E 1 Summary of previous sedimentological studies with a focus on the Sego Sandstone and their interpretations with respect to inferred depositional environments and processes.

Year
Author

| Description
Facies Association 1 is relatively common in the Sego Sandstone at the studied localities and is characterised by packages up to 6.5 m thick with concave-up erosional bases, lenticular architectures and overall fining-upward trends (Figure 4).This FA consists of fine to very coarse-grained sandstone and typically shows unidirectional tangential and planar-tabular cross-stratification, or less commonly is structureless or with horizontal planar parallel stratification.Palaeocurrents are typically unidirectional and mainly oriented towards the south.Dominantly unidirectional palaeocurrents have also been reported by previous authors and mainly show a south to south-east trend corresponding roughly to a palaeoseaward orientation (Legler et al., 2014;van Cappelle et al., 2016;Willis & Gabel, 2003).Some cross-stratification shows apparent cyclical patterns, highlighted by sparse organic or carbonaceous drapes, although these sedimentary structures are typically unidirectional and show continuous sedimentation rather than being systematically interrupted by reactivation surfaces (Figure 5A).Cyclical cross-stratification with fine-grained drapes has been reported in other studies from the Sego Sandstone and while locally more clearly organised than displayed in Figure 5, they mainly show similar features, such as unidirectional palaeocurrents, uninterrupted layers and mostly organic drapes (e.g.figure 5 of Legler et al., 2014;figure 5 of van Cappelle et al., 2016;figure 10 of Willis, 2000;figure 13 of York et al., 2011).At Crescent Canyon, FA 1 shows interbedding consisting of amalgamated decimetre-thick sandstone beds with rarer intervening heterolithic beds (Figure 6A).Sandstone beds show sharp or erosional bases, are commonly structureless or cross-stratified, and contain sparse mudstone rip-up clasts.The intervening beds consist of fine-grained sandstone with asymmetric rippled lenses, cross-lamination and mudstone drapes.Similar decimetre-thick interbedded units are also reported by Willis and Gabel (2001).Facies Association 1 is largely unburrowed, with sporadically distributed Ophiomorpha nodosa (Figure 7A) and borings of Teredolites in allochthonous wood (BI 0-1).

| Interpretation
Facies Association 1 largely corresponds to the interpreted tide-influenced to tide-dominated distributary channels of several previous studies (e.g.FA 7 of Legler et al., 2014;FA 4 of van Cappelle et al., 2016;FA 3 of Willis & Gabel, 2001).The basal erosional surfaces, fining-upward trends and depositional architectures suggest a channelised depositional element as also suggested by previous authors.Nonetheless, the process regime is herein revisited.Based on comparisons with recent literature (Dalrymple et al., 2015;Gugliotta et al., 2016b;Jablonski & Dalrymple, 2016;Sisulak & Dashtgard, 2012), the decimetre-thick alternations of erosionally based sandstone beds and heterolithic intervening deposits are interpreted as a preserved signature of river discharge fluctuations.
The sandstone beds are interpreted as river-flood deposits formed under fluvial-dominated and unidirectional conditions as also shown by their palaeocurrent trends.The heterolithic beds with mudstone drapes are interpreted as interflood deposits and show potential tidal influence evidenced by the asymmetric ripples and mudstone drapes.In addition, due to the unidirectionality, apparent continuous sedimentation and type of drapes, most of the potentially cyclic bedforms of the Sego are interpreted as the result of tidal modulation of a unidirectional riverine current (Gugliotta et al., 2016a;Hendershot et al., 2016;Martinius & Gowland, 2011), rather than as purely tidally generated bedforms (Visser, 1980).Furthermore, the high sandstone contents in FA 1 contrast with the much lower sand-mud ratios reported from modern active tide-dominated distributary channels (Dalrymple et al., 2003;Gugliotta et al., 2017Gugliotta et al., , 2019;;Wang et al., 2009) and with the known tidal  dynamics that help retain mud within these system (Gugliotta & Saito, 2019;Nowacki et al., 2015;Wolanski et al., 1995Wolanski et al., , 1996)).This suggests that the Sego channels likely had a highly efficient mechanism for exporting mud to the shelf, contributing to the formation of the Mancos Shale.It is suggested here that this mechanism was strong riverine dominance, with the relatively weak tidal processes unable to retain significant amounts of mud in the active channels.The uncommon and sporadically distributed trace fossils are interpreted to record temporary incursions of saline water into the channels, but this does not necessarily imply tidal dominance nor strong tidal processes.
Based on the predominant unidirectional and palaeoseaward orientation of these palaeocurrents, the high sandstone-mudstone ratios and the process-regime reinterpretation discussed above, the deposits of FA 1 are herein reinterpreted as fluvial-dominated distributary channels with subordinate tidal influence (i.e.fluvial dominated, tide influenced = Ft).

| Description
Facies Association 2 makes up the highest proportion of deposits in the lower Sego Sandstone at the studied localities (note their abundance in Figure 4).These deposits form up to 5 m thick coarsening and thickeningupward packages with slight erosional bases.They internally consist of a characteristic centimetre to decimetre-thick, semi-regular alternation of sandstone and intervening heterolithic beds (Figure 5B).This interbedding motif is present not only in this dataset but it is apparently common throughout the lower and upper Sego Sandstone in different parts of the Book Cliffs in Utah and Colorado, as observed in photographs from previous studies (e.g.figure 7 of Franczyk, 1989; figure 3  thick interval at Tusher Canyon, which consists of finegrained sandstone with common mudstone drapes and bidirectionally oriented asymmetric ripple lenses and cross-lamination (Figure 5C), is this interbedded pattern apparently absent.Overall, FA 2 is sandstone dominated, as also reported by van Cappelle et al. ( 2016).
In the characteristic interbedding, sandstone beds are fine to medium grained with sharp or slightly erosional bases, are commonly structureless or crossstratified, and show sparse mudstone rip-up clasts.As reported by Willis and Gabel (2001), these beds are typically inclined at 2-6° towards the south-east (i.e.2014) reports them to have oblique accretion towards the south-west.Sandstone beds become thicker and show more pronounced erosional bases towards the upper part of the packages.Palaeocurrents are dominantly unidirectional and mainly oriented seaward towards the south to south-east, as also reported by previous studies (Legler et al., 2014;Willis & Gabel, 2001).Locally, potentially cyclic bedforms with fine-grained drapes with the same characteristics as those of FA 1 are also observed.Willis & Gabel, 2001).The coarseningupward trends, inclined bedding and general architecture suggest a barform element, as also suggested by previous authors.Nonetheless, as for FA 1, the dominant decimetre-thick alternations of erosionally based sandstone beds and intervening heterolithic deposits are interpreted as the preserved signature of river discharge fluctuations associated with floods (sandstone beds) alternating with interflood intervals (heterolithic beds) (Dalrymple et al., 2015;Gugliotta et al., 2016b).The recognition of this widespread signature provides new elements for the reinterpretation of the process regime as dominantly fluvial.
Evidence for tidal and subordinate wave processes, such as heterolithic bedding with bidirectional and symmetric ripples, respectively, are largely restricted to the interpreted interflood beds formed when river discharge was low.It is suggested that these marine processes were inefficient in reworking the interbedded pattern generated by the riverine currents.The trace-fossil assemblage is attributable to the Rosselia Ichnofacies (MacEachern & Bann, 2020), but high intensity and high diversity conditions are restricted to the interpreted interflood beds, suggesting that saline conditions were restricted to these periods and largely controlled by the riverine fluctuations.Similar to FA 1, the potentially cyclic bedforms are interpreted to be the result of tidal modulation of unidirectional riverine currents (Gugliotta et al., 2016a;Hendershot et al., 2016;Martinius & Gowland, 2011).Owing to the dominant interbedded signature in conjunction with the unidirectional and palaeoseaward -oriented palaeocurrent pattern, and high sandstone-mudstone ratio, FA 2 is reinterpreted to record fluvial-dominated mouth-bar deposits with a subordinate tidal influence and limited wave process (i.e.fluvial dominated, tide influenced, wave affected = Ftw).

| Description
Facies Association 3 comprises the highest proportion of the Buck Tongue at any of the studied localities (note abundance in Figure 4).The deposits form packages up to 6 m thick with coarsening-upward trends internally composed of centimetre to decimetre-thick, semi-regular alternations of sandstone beds and intervening heterolithic or mudstone beds (Figure 5D).Overall, FA 3 shows nearly equal proportions of sandstone and mudstone.The bedding appears sub-horizontal at outcrop scale and displays continuous lateral extension for at least hundreds of metres to kilometres, as noted previously by Legler et al. (2014) and Willis and Gabel (2001).In the characteristic interbedding, sandstone beds consist of very fine to fine-grained sandstone with sharp or slightly erosional bases and, locally, symmetrically reworked tops.Sandstone beds may be structureless or show symmetric ripple lamination or hummocky cross-stratification (Figure 8A,B,C).The intervening mudstone beds contain sandstone layers in the form of millimetre-scale symmetric and asymmetric ripple lenses and cross-lamination, as well as syneresis cracks.Rare asymmetric ripples show palaeocurrent orientations towards the south-east (i.e.palaeoseaward).

| Interpretation
Facies Association 3 largely corresponds to the interpreted wave-dominated delta front and/or prodelta in several previous studies (e.g.FA 1-3 of Legler et al., 2014; FA 1 of van Cappelle et al., 2016; FA 1 of Willis & Gabel, 2001).The coarsening-upward trends, sub-horizontal bedding and lateral extension of the beds suggest an unconfined setting consistent with a deltafront interpretation.The decimetre-thick alternations of erosionally based sandstone beds and heterolithic intervening deposits, which are dominant in this FA, show sedimentary structures (e.g.hummocky crossstratification, symmetric ripples) that are typically associated with fair-weather and storm deposition within wave-dominated environments (Onyenanu et al., 2018;Vakarelov et al., 2012;van Cappelle et al., 2016).Nonetheless, similar structures can also form in distal zones of more fluvial-dominated deltas (Allison et al., 2005).
In modern wave-dominated environments, the interbedded signature is locally observed but lacks the vertical and lateral continuity reported in FA 3, probably due to more persistent wave reworking and biological activity in conjunction with relatively low sediment supply (Khan et al., 2021;Qiaola et al., 2022;Tamura et al., 2008).
In agreement with Bhattacharya et al. (2020), it is suggested here that the persistent interbedding pattern of FA 3 should be associated with high sediment supply near a river mouth and therefore to a riverine signature.
In FA 3, wave processes reworked individual beds but without fully overprinting the general interbedding pattern attributable to combined riverine-wave processes.
The trace-fossil suite, attributable to the Rosselia Ichnofacies (MacEachern & Bann, 2020) and the variability of BI within the interbedded pattern reflects the temporal variations in physico-chemical stresses, consistent with a mixed riverine-wave signature.Furthermore, the lateral and vertical facies relationships must also be taken into account a reliable interpretation.If assuming a genetic link between the Tongue and the overlying Sego Sandstone, as suggested by Legler et al. (2014) and van Cappelle et al. (2016), it becomes more logical to interpret FA 3 as the distal expression of FA 2, rather than as a significantly different depositional environment.If wave processes dominated the distal delta front, the shoreline (i.e.FA 2) should show much stronger evidence of wave reworking.The interbedded motif of FA 3 is interpreted as the result of river floods combined with fair-weather and storm wave reworking that alternated with periods of ambient marine conditions, recording a cryptic and distal expression of the river-derived interbedding interpreted in FA 2. The deposits of FA 3 are interpreted as a distal delta-front environment and considered overall to be fluvial dominated, wave influenced and tide affected (= Fwt), although a wave-dominated, fluvial-influenced and tide-affected (= Wft) interpretation at the local level is not fully discarded.

| Description
Facies Association 4 is typically intercalated with FA 3 and forms a subordinate component of the Buck Tongue (Figure 4).The deposits consist of very fine to fine-grained sandstone with hummocky cross-stratification with sparse mudstone clasts, forming beds or bedsets up to 1 m thick.

| Interpretation
Facies Association 4 largely corresponds to FA 2 of Legler et al. (2014), FA 2 of van Cappelle et al. ( 2016) and FA 1 of Willis and Gabel (2001).Facies Association 4 represents distal delta-front deposits that were, in some cases, thoroughly reworked by storm waves that likely removed most intervening mud and potential riverine interbedding.The presence of intercalated mudstone rip-up clasts likely records the erosional removal of the intervening heterolithic deposits.Facies Association 4 is interpreted as wave dominated (W).Nonetheless, when considering the distribution of this FA, the wave-dominated environment should be considered as relatively localised and temporary, rather than expressing the main signature in the delta front.

| Description
Facies Association 5 is quite rare in the studied intervals (Figure 4).It consists of structureless to laminated mudstone with millimetre-scale, very fine to fine-grained sandstone laminae and lenses.Sporadically distributed bioturbation of variable intensity (BI 0-3) prevails, with ichnological suites consisting of Asterosoma, Chondrites, Lockeia, Phycosiphon, Planolites, Teichichnus and Thalassinoides.A thin layer with oyster shells is found in this FA at Tusher Canyon.

| Interpretation
Facies Association 5 is interpreted as a quieter depositional environment in the distal area of the deltaic system, such as the prodelta.The trace-fossil suite is attributable to the Phycosiphon Ichnofacies (MacEachern & Bann, 2020).

| Widespread riverine signature and reassessment of the process regime
The deposits of the Sego Sandstone/Buck Tongue system are commonly considered a typical example of a tide or mixed tide-dominated and wave-dominated palaeoenvironment (Table 1).In this study, interpretable tidal and wave generated structures, as well as trace-fossil suites highlighting marine conditions are observed in the studied units.Although their significance is not underestimated, the mere presence of the indicators of a particular process does not necessarily make the case for its dominance or strong influence in the palaeoenvironment.In addition, other non-marine process signatures are also common but were not previously analysed in detail.The studied outcrop locations are those that were central in several previous studies (van Cappelle et al., 2016;Willis & Gabel, 2001, 2003).The reviewed literature also suggests that the sedimentary features reported herein are not only present locally but are widespread and persistent in the studied units across the entire Book Cliffs.The primary observations of the sedimentary facies, FAs and architectures in the studied units and the resultant interpretations of the deltaic depositional elements (e.g.channels, bars), are not dissimilar to those proposed in previous studies; however, one fundamental exception exists in the observations surrounding the interbedded facies motif that gives rise to major differences in the process interpretation.
The studied show a widespread interbedded pattern that was largely overlooked or underemphasised in past, but which needs to be carefully considered as it provides, with reference to analogous examples in both ancient and modern coastal to shallow-marine deposits, new evidence for reassessing the process regime.In the Sego Sandstone (the primary subject of this study), this signature is particularly evident (Figure 6) and is interpreted to record fluctuations in river discharge associated with alternating flood and interflood periods.The description of these interbedded intervals and their interpretation as a riverine signature are in line with a large body of recent literature derived from several other rock units (Collins et al., 2017;Dalrymple et al., 2015;Gugliotta et al., 2016b;Jablonski & Dalrymple, 2016;van Yperen et al., 2020), as well as from modern systems (Gugliotta et al., 2018;Sisulak & Dashtgard, 2012).Similar interpretations of such facies relationships had also been made decades earlier (Thomas et al., 1987).In agreement with this recent literature, it is believed that the interbedded motif of the Sego Sandstone with up to decimetre-scale beds is difficult to explain by tidal processes alone.In fact, tidal rhythmites typically show thinner laminae, generally residing at the millimetre to centimetre-scale and showing more gradual transitions between sandier and muddier intervals associated with neap-spring or other cycles (Gugliotta et al., 2022;Longhitano et al., 2012).In deposits of the distal delta front of the Buck Tongue (the secondary target of the study), the pattern is more cryptic, owing to stronger reworking by waves and more common presence of associated wave-generated sedimentary structures.Nonetheless, it appears that the river signature is still preserved and represented by the interbedded pattern (Figure 8), wherein sandstone beds correspond to river-flood events combined with storms and/or fairweather wave reworking (Collins et al., 2017;MacEachern et al., 2005).This reinterpretation is further supported by the ichnological suites in both the Sego Sandstone and Buck Tongue, with trace fossils largely restricted to the interflood beds (Figures 6,7,8), indicating the prevalence of marine conditions during these periods (MacEachern et al., 2005;MacEachern & Bann, 2020).The sandstone beds, however, remain largely unburrowed or display rare, top-down bioturbation, consistent rapid event deposition and physico-chemically stressed conditions, likely related to strong river processes.In contrast, modern tidedominated deltas show more persistent tidal and saline conditions throughout the year, at least near their river mouths and within a portion of their channels, resulting in more pervasive presence of marine facies and/or bioturbations (Dalrymple et al., 2003;Gugliotta et al., 2017Gugliotta et al., , 2022;;Wolanski et al., 1995Wolanski et al., , 1996)).
In the studied units, waves and tides were ineffective in reworking the signature of riverine processes, except for some localised areas, despite operating more frequently compared to flood-interflood fluctuations.By contrast, Brown and Hale (2021) report that in deposits of the modern tide-dominated Ganges-Brahmaputra delta, regular reworking by tides obliterates any potential signatures of seasonal riverine variations.Similar observations were also made by Gugliotta et al. (2022) for the tide-dominated and wave-dominated Mekong delta, where the monsoonal signature is thoroughly overprinted by tidal processes in the tide-dominated part of the system.The deposits of modern wave-dominated deltas and strandplains also show more intense marine reworking and reduced preservation of interbedded patterns, although some of these beds are present locally (Khan et al., 2021;Qiaola et al., 2022;Tamura et al., 2008).It is therefore suggested that the widespread preservation of the interbedded riverine signature from interpretable delta plain to delta front environments in the Sego Sandstone/Buck Tongue system is an indication of rather weak reworking by marine processes.In the full spectrum of shallow-marine deposits from purely fluvial to strongly marine reworked, it is interpreted that the deposits of the Sego Sandstone and Buck Tongue fall in an intermediate category wherein the river signature is still clearly preserved, likely representing the dominant process, but with evidence of subordinate wave and tidal signatures also preserved, indicating a mixed-energy setting (Figure 9).In the context of existing mixed-energy interpretations for the Sego Sandstone, the revised fluvialdominated, tide-influenced and wave-influenced interpretation presented here represents an evolution within the continuum of coastal to shallow-marine processes.
In addition to the type of facies preserved, the position in the depositional system where a specific process signature is preserved also provides important information about the overall process regime.Typically, waves are restricted to the shoreline and only penetrate a limited distance landward within channels during storms.Tides can extend for up to hundreds of kilometres along the FMTZ (Gugliotta et al., 2017(Gugliotta et al., , 2019)).This implies that the location where predominantly river signatures are preserved versus where mainly tidal signatures are recorded, as well as in which sub-environment (channel, shoreline, delta front, etc.) these occur, have important implications for the overall process-regime interpretation of the system.For example, a similar interbedded motif as the one reported in this study was described from point-bar deposits of the McMurray Formation, which are interpreted as fluvial dominated and tide influenced (Jablonski & Dalrymple, 2016).In that case study, this signature is believed to have formed relatively upstream, suggesting that hypothetical (but time-equivalent deposits nearer the river mouth were likely to been tide (or tide-wave) dominated (Figure 10A).Conversely, in this case study from the Book Cliffs, the fluvial-dominated, tide-influenced and wave-influenced signature occurs in deposits of the delta front, suggesting comparatively weak marine processes and reduced influence of marine processes in the palaeolandward direction (Figure 10B).The middle Castlegate Sandstone preserves the contemporaneous fluvial part of the Sego Sandstone/Buck Tongue system (Figure 1B) and shows no significant tidal processes recorded in its facies, as would be expected for a fluvial-dominated, tide-influenced and wave-influenced deltaic shoreline.The FMTZ of the Sego Sandstone/Buck Tongue system was likely much shorter compared to the one in the Mekong delta, where tides dominate for about 100 km from the shoreline and affect the hydrodynamics for at least another 200 km inland (Gugliotta et al., 2017).Similar spatial considerations can also be made for the wave processes.Modern wave-dominated systems show a wave dominance at their shorelines and across the delta front, whereas the Sego Sandstone/Buck Tongue system shows significant wave reworking only in the distal delta front.This proximal-distal relationship in the studied units can also be seen in modern fluvial-dominated systems such as the Mississippi River delta (Allison et al., 2005), and would be more difficult to explain with a wave-dominated model.Despite the importance of understanding the proximal to distal and lateral variations of depositional processes, only a few studies make similar considerations when assessing the process regime from the rock record (this study; Gugliotta et al., 2016a;2016b;van Yperen et al., 2020).

| Towards a more comprehensive assessment of sedimentary records
In terms of the overall depositional environment, the present study agrees with previous works that suggest a deltaic interpretation for the studied units (Legler et al., 2014;van Cappelle et al., 2016;Willis & Gabel, 2001, 2003).Conversely, an interpretation as estuarine or barrier system for the entire interval would be problematic when considering the interbedding patterns, thicknesses of FAs and units, high sandstone-mudstone ratios, presence of coarseningupward trends and overall progradational character of the interval.That said, some transgressive and/or nondeltaic deposits may be present locally.The previous views regarding a particularly strong control of tidal process or of a mixed tidal and wave process regime with only subordinate riverine processes in the deltaic system are challenged here.Integrating the detailed observations of sedimentology, such interbedding arrangements, vertical sandstone-mudstone ratios, palaeocurrents patterns, relationships, ichnology and stratigraphic architecture (including the stratigraphic relationships with adjacent age-equivalent formations) and comparing these to other ancient analogues with comparable features, provides the basis for reliable process-regime interpretation (Figure 11).When integrating this range of observations beyond the commonly employed process indicators and the interpretation of erosional surfaces as representing tidal channel erosion, the reinterpretation as mixedenergy, fluvial-dominated delta perhaps better explains features that appeared problematic with previous interpretations.For example, the predominant unidirectional, palaeoseaward orientation of measured palaeocurrents, which was originally attributed to ebb-tide dominance (Willis & Gabel, 2001, 2003), contrasts with the common bidirectionality of tide-dominated systems, even when assuming mutually evasive patterns.It was also reported that the palaeoseaward-dipping bedding of the interpreted tidal bars was different from the ones of modern tidal bars, and instead were more similar to the mouth bars of fluvialdominated deltas or inlet deltas (see p. 503 of Willis & Gabel, 2001).Furthermore, the high sandstone-mudstone ratios in the Sego Sandstone contrast with the much lower sand-mud ratios the typify the deposits of modern tide or mixed tide-dominated and wave-dominated deltas, such as the Mekong River delta (Gugliotta et al., 2017(Gugliotta et al., , 2019(Gugliotta et al., , 2022)), which has been proposed as an analogue for the studied units (Legler et al., 2014;van Cappelle et al., 2016).The reinterpretation of the Sego Sandstone/Buck Tongue system as a mixed process, fluvial-dominated delta would better explain the accretion and palaeocurrent patterns observed and would reconcile the discrepancy regarding the sandmud ratios.Furthermore, erosional surfaces interpreted as incision of valleys (van Wagoner, 1991a(van Wagoner, , 1991b) and more lately as scour at the bases of tidal channels (Legler et al., 2014;van Cappelle et al., 2016;Willis & Gabel, 2001, 2003), could also be explained by autogenic erosion of distributary channels and/or fluvial processes at the delta front (Fielding et al., 2005).Weaker tidal processes would also be consistent with regional basin modelling of the Cretaceous WIS, which suggests that microtidal conditions and weak bed shear stresses associated with tidal currents were likely unable to transport sand in the Utah Bight (Dean et al., 2019;Ericksen & Slingerland, 1990).Although Dean et al. (2019) suggested that local confinements could explain the commonly reported tidal deposits in the Sego Sandstone, the apparent discrepancy between the modelling and field observations is also easily explained by reconsidering previous interpretations.The overlying Neslen Formation, which was previously interpreted as a tidal environment (Steel et al., 2012;Willis, 2000), has also been recently reinterpreted as a mixed-energy, fluvialdominated palaeosystem (Shiers et al., 2014(Shiers et al., , 2017) ) that contains a signature of varying river discharge (Dalrymple et al., 2015).
Upon a review of the existing literature on the Sego Sandstone, additional doubts arise regarding its interpretation as the product of a strong tidal system.The potentially cyclic bedforms present in the Sego Sandstone, as reported in Figure 5A and shown in other papers, historically have been considered indicative of tidal process, following studies from Holocene tidal environments (Visser, 1980).Nonetheless, a large and growing body of literature (Ainsworth et al., 2012;Gugliotta et al., 2016a;Hendershot et al., 2016;Martinius & Gowland, 2011;Reesink & Bridge, 2011) provide alternative and probably more suitable explanations for these sedimentary structures and needs to be more carefully considered.The majority of the Sego Sandstone bedforms show drapes consisting of organic material and show apparently continuous sedimentation rather than recurring and rhythmic interruptions in bedform migration marked by reactivation surfaces.Martinius and Gowland (2011) suggested that the drapes of these bedforms do not require slackwater conditions to form and are likely the result of minor fluctuations in unidirectional currents.As a result, some of these bedforms may have formed under tidal modulation of riverine currents, and therefore represent fluvial-dominated, tideinfluenced signatures (Gugliotta et al., 2016a;Hendershot et al., 2016;Martinius & Gowland, 2011).In addition, similar cyclic and draped bedforms can also form in nontidal environments, due to the dynamics of superimposed bedforms or wind action (Ainsworth et al., 2012;Reesink & Bridge, 2011).Consideration of such studies are crucial for reaching more refined interpretations of coastal to shallow-marine systems, particularly in avoiding the over-interpretation of tidal processes and not considering alternative process models.When considering all the sedimentological, ichnoand stratigraphic observations together its regional depositional context, the Sandstone/Buck Tongue system is better explained using a mixed-energy but fluvial-dominated deltaic model (Figure 12).Although wave and tidal reworking likely played a role in its deposition, certainly on a local scale, it is likely that both marine processes were subordinate to fluvial processes, with tides potentially being the weakest of the three.However, the studied system remains different from typical fluvial-dominated deltaic units of the WIS, such as the Ferron Sandstone or the Panther Tongue, where marine processes were even weaker (Ahmed et al., 2014;Enge et al., 2010).The tendency to put more emphasis on differences (e.g.drapes) rather than similarities (e.g.architecture, sandstone-mudstone ratios, interbedding patterns) has led to, and ultimately reinforced, the tendency of interpreting relative end members (e.g.fluvial versus tide dominated).These findings highlight a potentially significant issue with the established approach of assessing process regimes based on the selective identification of certain process indicators with entrenched process interpretations.This commonly results in other important sedimentological and ichnological observations and alternative views being inadvertently overlooked or underemphasised.In the Sego Sandstone and possibly for many other similar units there has been a tendency to emphasise marine processes and to disregard or diminish the role of riverine currents.The prevailing approach of recognising relative process end members has provided an important framework that enables gross differences between units to be recognised and compared.However, as knowledge has widened and deepened, this simplistic framework is clearly unsuitable for capturing the complexity of both modern and ancient coastal to shallow-marine systems.To achieve refined, high-resolution interpretations of these systems, continued development of mixed-process F I G U R E 1 2 Depositional model for the lower Sego Sandstone/Buck Tongue system, which was interpreted as a fluvial-dominated, wave-influenced and tide-influenced deltaic system.Note the distributions of FA 1 through 5, the inferred process regime (e.g.F, Ft, Fwt) following the nomenclature of Ainsworth et al. (2011), and the distribution of the interpreted Fluvial to Marine Transition Zone (FMTZ) of this system.models that employ and integrated assessments the sedimentary record is essential.

| CONCLUSIONS
This study reevaluates the preserved process regime of the Cretaceous Sego Sandstone/Buck Tongue system in the Book Cliffs.The findings reveal a previously overlooked or underemphasised fluvial-dominated signature in the deposits, which consist of alternating centimetre to decimetre-thick sandstone and heterolithic beds, strongly suggestive of variable river discharge during flood-interflood periods.The recognition of a widespread riverine signature and reinterpretation of these units as a fluvial-dominated system are in line with recent literature and highlight the broader need for a change in approach to the interpretation of the sedimentary records.This study highlights the need to assess the entire dataset comprehensively across all scales, with reference to the relationships with other age-equivalent formations and the geological setting.High-resolution studies must consider and evaluate multiple hypotheses to decipher the actual complexity of ancient coastal to shallow-marine systems.Additional work is needed to fully understand the deposits of the Sego Sandstone/ Buck Tongue system, as well as other similar rocks, but first requires the development and refinement of an improved approach.

ACKNO WLE DGE MENTS
We thank the journal editor Peter Swart and reviewers Alessandro Amorosi and Nigel Mountney for their valuable comments, which have contributed to significant improvements to this manuscript.Anna van Yperen and Ernesto Schwarz are thanked for their comments on a previous version of this manuscript and David Hodgson and Stephen Flint are thanked for introducing Marcello Gugliotta to some of the outcrops in the Book Cliffs.

F
Palaeogeographical reconstruction of the Western Interior Seaway in the middle Campanian, showing the location of the Utah Bight where the Sego Sandstone accumulated (© 2022 Colorado Plateau Geosystems Inc.).

F
I G U R E 3 Panoramic outcrop view of the Buck Tongue and lower Sego Sandstone at Thompson Canyon, showing a general coarsening-upward trend and relatively gradational contact.

F
I G U R E 4 (A through C) Sedimentological logs of the Buck Tongue and lower Sego Sandstone at three studied localities along the Book Cliffs in Utah showing the vertical distribution of the facies associations and their characteristics.See Figure 1A for locations.
of Legler et al., 2014; figure 4 of van  Cappelle et al., 2016; figure 10 of Willis, 2000; figure11ofYork et al., 2011).Only in an exceptional 0.6 m F I G U R E 5 Selected photographs of sedimentary facies.(A) FA 1 showing cross-stratified sandstone with sporadically distributed carbonaceous drapes.(B) Coarsening-upward trend in a package of FA 2, showing its characteristic alternation of sandstone and heterolithic beds.(C) Detail of FA 2 with asymmetric ripple lenses and mudstone drapes in the strongly reworked atypical deposits reported at Tusher Canyon.(D) FA 3 showing its characteristic alternation of sandstone beds and heterolithic beds.U R E 6 (A through H) Details of interbedded intervals in FA 1 and 2 of the lower Sego Sandstone, showing the characteristic centimetre to decimetre-scale alternation of sandstone beds (interpreted as river-flood deposits) and heterolithic beds (interpreted as interflood deposits).Sandstone beds are generally structureless or cross-stratified and contain mudstone rip-up clasts.By contrast, heterolithic beds contain mudstone drapes, marine trace-fossil suites and asymmetric ripple lenses.Thickness of each bed varies from a few centimetres to several decimetres.Contacts are typically irregular to sharp at the bases of sandstone beds and relatively more gradational at the bases of heterolithic beds.In some cases (e.g.A and E), sandstone beds are amalgamated and little or no heterolithic deposits are present.

F
I G U R E 8 (A through C) Details of interbedded pattern in FA 3 of the Buck Tongue, showing the characteristic centimetre to decimetre-thick alternations of sandstone beds (interpreted as river floods combined with wave reworking) and heterolithic beds (interpreted as interflood deposits).Sandstone beds are generally structureless or hummocky cross-stratified and show symmetric ripple lamination and oscillation reworking at the top.By contrast, heterolithic beds consist of lenticular bedding with ripple lenses and bioturbation.The thickness of each bed varies from a few centimetres to several decimetres.Contacts are typically irregular to sharp at the bases of sandstone beds, and sharp to gradational at the bases of heterolithic beds.

FB
Conceptual examples of different degrees of preservation of river-generated interbedding versus marine reworking and interpretation of the deposits of the Sego Sandstone/Buck Tongue system.Conceptual model highlighting the importance of the location in the system where fluvial-dominated, tideinfluenced facies (Ft) are preserved in relation to the development of the Fluvial to Marine Transition Zone (FMTZ).(A) A potentialy tide dominated system where Ft facies are found upstream (e.g., McMurray Formation).(B) A mixed-energy, fluvial-dominated system where similar Ft Facies are found comparatively downstream (e.g., Sego Sandstone/Buck Tongue).A Ft facies found upstream (e.g., McMurray Formation) Ft facies found downstream (e.g., Sego Sandstone/Buck Tongue) FMTZ (Wtf or Twf) T-dominated, f-influenced (Tf) F-dominated, t-influenced (Ft)

F
Conceptual sketch showing the scope and relative weighting of observations employed for process-regime interpretations of the Sego Sandstone.
Map of study area in the Book Cliffs of Utah (USA).(B) Stratigraphy of the Campanian (Upper Cretaceous) Mesaverde Group in Utah and Colorado (modified from van Cappelle et al., 2016).