Predictive facies and sequence-stratigraphic models (Hart 2015, 2016) for fine-grained sediment deposition associated with a peak transgression in an epeiric seaway have been tested here using elemental, mineralogical, TOC, and wireline log data. Our samples were collected from close to the Cenomanian–Turonian boundary from five widely dispersed locations (Texas, New Mexico, Colorado, Alberta, Manitoba). We collected data from a range of lithotypes (shales, marlstones, and limestones) from both core and exposures to test the models. Profiles of major elements (Si, Al, Ca, etc.) and TOC show distinctive trends depending on the relative proximity of the location to the shoreline at peak transgression. Siliciclastic minerals (and hence elements such as Si, Al, K) dominate the section in our most proximal location whereas peak transgression was dominated by deposition of pelagic carbonates in distal locations. Organic-carbon content is variably related to sediment composition, again depending on the relative proximity of the location to the shoreline at peak transgression. TOC is highest in marlstones (deposited where pelagic and siliciclastic sediments mixed) and decreases both in basin-center limestones and in proximal, siliciclastic shales.

The Early Cretaceous Robberg Formation is a clastic, syn-tectonic unit deposited near the updip limit of the Pletmos Basin during initial separation of Africa and South America. Robberg sandstones were deposited in estuarine tidal channels whereas siltstones fill abandoned channels. Abundant soft-sediment deformation and dewatering structures have previously been attributed to earthquakes. Our re-examination of a part of the Robberg Formation confirmed the presence of load casts due to inverse density stratification, whereas sand dikes probably record seismically triggered dewatering. Layers of siltstone pervasively disrupted by microfaults are probably seismogenic and matrix-supported, intraclastic breccia is interpreted to record seismogenic liquefaction and channel-bank collapse. However, the most common types of soft-sediment deformation, seen mainly in vertical cross-section and spatially and stratigraphically isolated in undeformed sediment, are attributed to dinosaurs. Track preservation takes five main types: (1) Undeformed to obliquely stratified sandstone fills sharply defined track impressions in siltstone in which lamination is truncated and deformed downward. (2) Bowl-shaped depressions, 50 cm to >1 m in diameter, in heterolithic channel-plug facies are transmitted undertracks, probably produced by sauropods. (3) Dinosaurs walking on a thin sand layer on the bottom of a tidal channel punched the sand down into underlying channel-plug siltstone, the latter being highly deformed. (4) Dinosaurs walked or waded in the mud of abandoned channels, producing highly deformed, destratified and sheared sediment, surrounded by wholly undeformed sediment. (5) Dinosaurs walked on moist sandy channel point-bars producing locally deformed, brecciated, and de-stratified, bowl-shaped depressions, bordered by folds and thrusts. The majority of tracks appear to be broadly cylindrical, and lack obvious digits, which together with size, suggest a sauropod (or possibly other quadruped) origin. A few tracks, seen in plan, have three digits, suggesting a bipedal trackmaker. This track assemblage appears to be the first recognized in the Cretaceous of southern Africa.

Late Pleistocene coastal aeolian sandstones, characterized by metre- to tens of metre-scale cross-stratification and dated to late Marine Isotope Stage 6 or early Stage 5 (153 to 114 ka), are exposed on the Cape coast of South Africa. At four localities, upward-radiating burrow systems extend vertically through as much as 2.7 m of aeolian dune sandstone and span over 1 m horizontally. Burrows are cylindrical, 2-4 cm in diameter, and unlined. Burrows radiate subtly upward, but do not appear to branch. There is no central chamber at the apex of the burrow system. One burrow system in aeolian sandstone immediately overlies sandstone with marine invertebrates and Ophiomorpha, implying proximity to a saline water table, whereas the other burrow systems are hosted entirely in non-marine, aeolian facies. The burrow systems are typically spatially associated with up-arching of stratification in the enclosing aeolian dune sandstone. This geometry implies that the burrows, or burrowing organisms, interacted with sedimentation to cause preferential accumulation of sand around the burrow system. Possible explanations include trapping of sand around projecting burrow turrets, up-building due to excavated sand or sand adherence to a damp surface. The features of the burrow systems are incompatible with those of buried plants. The burrow systems resemble 'cluster burrows' described from Holocene-Pleistocene Bahamian aeolianites, interpreted as escape burrows made by emerging juvenile wasps. However, escaping juvenile wasps would not cause up-arched lamination. The burrows do not resemble those known to be made by ants, termites, beetles, wasps, bees or decapod crustaceans. Burrowing spiders appear to make structures most closely resembling the burrows that we observe. The traces are assigned to Lockleyichnus gerickensis, igen. et isp. nov.

A single trackway of the heron-like ichnogenus Ardeipeda sp. is described from alluvial plain sediments of the Cretaceous (middle Cenomanian; 96-97 Ma) Dunvegan Formation, British Columbia. This is both the largest avian footprint known from the Cretaceous, and the oldest (by >45 myr), report of the ichnogenus Ardeipeda. It cannot be proved or disproved that the trackmaker represented the extant heron family (Ardeidae); no body fossils of the family are known before the early Oligocene. Thus, it is inferred that the trackmaker was ecologically, rather than biotaxonomically, convergent with modern herons. This conclusion is consistent with growing evidence of convergence between Cretaceous, Cenozoic and extant avian track morphologies.

We compare the Second White Specks and Belle Fourche For-mations (2WS/BF) of Alberta, Canada to the Eagle Ford Forma-tion shale (EF) of southern Texas and, to a lesser extent, theTuscaloosa marine shale (TMS) of Louisiana and Mississippi.These units were deposited during the same Late Cretaceous(Cenomanian-Turonian) eustatic rise and highstand, and allare prolific source rocks. Our aim was to define the reasonswhy attempts to develop the 2WS/BF as a resource play wereunsuccessful. We propose that failure was controlled by thecombined effects of three geologic factors-reservoir quality,organic richness, and completion quality-all of which stemfrom the depositional setting and history of the interval. Becauseit was deposited in close proximity to a major source offine-grainedsiliciclastic sediments, the 2WS/BF has a higher clay-mineral con-tent than the EF, the latter being deposited in a relatively clastics-starved setting. Porosity loss, through compaction, was thereforegreater in the 2WS/BF, and the higher clay-mineral content ofthat play, combined with the stratigraphic stacking, led to com-pletion problems such as proppant embedment and out-of-zonegrowth of hydraulic fractures. The TMS was likewise challengedas a resource play because of its high clay content. The originalorganic content of the 2WS/BF was neither as high nor as oilprone as that of the EF because of paleoenvironmental conditionsat the seafloor at the time of deposition. This retrospective analy-sis identifies depositional setting as afirst-order control on thepotential of a source rock to form a resource play.

Marine mudstone of Coniacian age (c. 89.51–86.49 Ma) was deposited on a storm-dominated ramp spanning the foredeep of the Cretaceous Western Canada Foreland Basin. Marine flooding surfaces define 18 allomembers that thin over 300 km, from c. 140 m in the proximal foredeep to c. 20 m close to the forebulge crest. The broadly conformable succession of allomembers is partitioned into five ‘tectono-stratigraphic units’ by low-angle unconformities that bevel off c. 10 to 20 m of strata over ‘arches’ that have a length scale of c. 50–100 km and are bounded by relatively linear zones of flexure. Depositional history involved two alternate modes: ‘Background’ deposition of subtly-tapered allomembers took place on a planar sea floor, subject to regional flexural subsidence, with sea-level modulated by Milankovitch-scale (c. 125 kyr) eustatic cycles. ‘Flexural’ events deformed the strata into troughs and arches across narrow zones of flexure. Arch crests were bevelled off, probably by submarine wave erosion. Eroded sediment did not accumulate in troughs but was advected beyond the study area by storm-driven processes. Cycles of deposition, warping and erosion were repeated five times on an average timescale of 600 kyr. Arches and troughs do not coincide with Precambrian basement structures, and their origin remains enigmatic. Changes in in-plane stress may have effected the localized vertical motion.

Delta-plain strata of the Cenomanian Dunvegan Formation, in north-east British Columbia, represent anastomosed rivers that were flanked by vegetated crevasse splays, wetlands and shallow lakes. These rocks preserve a rich record of ankylosaur and ornithopod walking and wading tracks, as well as crocodylian swim traces and tracks. Analysis of depositional, erosional, and trace-making events reveals alternating phases of flooding and emergence that controlled both animal activity (walking, wading, swimming) and mud substrate consistency (firm vs. soupy), and hence fidelity of track preservation. Some tracks resemble those produced by both ankylosaurs and crocodylians, and confident attribution is difficult. The absence of tail drag marks suggests an ankylosaur trackmaker, whereas prominent claw marks might favour a crocodylian origin. Unusually large crocodylian swim traces suggest animals about 9 m, and possibly up to 12 m in length. These large traces may represent a precursor to the giant crocodylian Deinosuchus. Ankylosaur tracks are well-preserved in lake and channel-margin deposits, whereas crocodylian traces are most commonly preserved in firm muddy lake sediments that had experienced previous subaerial exposure and dewatering. Close integration of trace fossils with sedimentary and stratigraphic features makes it possible to reveal a detailed chronology of biotic and sedimentary events on palaeo-surfaces.

The Cenomanian–Turonian Oceanic Anoxic Event (OAE 2: 94 Ma) marked one of the most severe episodes of climate and environmental change during the Cretaceous Period. The primary feature of this interval was widespread marine anoxia/euxinia, leading to deposition of organic-rich, thinly-bedded, mudstones across the world’s oceans, which in turn caused a pronounced positive shift in δ13C of seawater that is recorded (and characterizes) OAE 2 strata worldwide. The event was also marked by climate warming (with superimposed cooling pulses), biotic stress, and terrestrial perturbations such as increased continental weathering. However, the majority of studied records of OAE 2 were deposited in deep-marine Atlantic and Tethyan settings or European epicontinental basins. Thus, the record of environmental change in other locations or environments is less clear. The Western Interior Seaway (WIS) represented a marine corridor across North America that connected the Arctic and Atlantic marine realms. Thus, understanding the environmental and oceanographic changes in the WIS during OAE 2 is crucial to resolving the wider impact of the event across the global marine realm. Several previous works have focused on sites towards the southern end of the WIS (e.g., Pueblo, Iona Core). In this study, we present a new multi-proxy geochemical dataset from a site in the central–northern part of the Western Interior Seaway: Pratts Landing (western Alberta, Canada). Previous palynological studies have highlighted a southward migration of boreal dinoflagellates during the Plenus Cold Event midway through OAE 2, as well as increased input of terrestrial organic matter. Here, we correlate these data with information from redox, nutrient, and volcanic proxies, and compare the Pratts Landing record with other deep- and shallow-marine records of OAE 2 to gain a wider perspective over the environmental changes that operated in different settings during that time interval. This viewpoint is key for understanding the differences and complexities in how surface phenomena were disturbed during OAE 2, and interpreting geochemical records of different settings during that time interval.

Abstract Controversy exists regarding the timing of emplacement of oil in the giant Athabasca and Peace River oil sands. Bitumen-cemented sandstones are present in the late Albian Paddy Member of the Peace River Formation; some cemented sandstones formed burrowed firmgrounds and reworked intraclasts, showing that oil was reaching surface by about 101–102 Ma. Estuarine and shallow-marine sandstones of the Paddy Member are exposed on the Peace and Heart rivers in the vicinity of the town of Peace River. Bitumen-cemented sandstone occurs in four stratigraphic settings: 1) An in-situ cement forms locally bedding-transgressive, sheet-like masses in cross-bedded estuarine sandstone; 2) A bitumen-cemented and heavily burrowed layer, 30–60 cm thick, lies immediately beneath a marine transgressive surface and shows that oil infiltrated downward into, and cemented, the upper surface of a shoreface sandbody. The firm sand was subsequently exhumed by transgressive erosion, fractured and burrowed by arthropods; 3) Rounded pebble- to cobble-sized clasts of bitumen-cemented sand lie on the floor of a tidal channel, and the channel floor surface is also stained and burrowed. This case suggests floating oil infiltrated the channel floor at low tide, subsequently hardening prior to erosion and burrowing; 4) Rounded cobble-to boulder-size clasts of bitumen-cemented sandstone, up to 1 m wide and 0.5 m thick, lie on a regional marine ravinement surface cut on the uppermost Paddy shoreface sandstone. The sand was permeated with oil, top-down, then oxidized to form a tough bitumen cement. The bituminous sand was subsequently scoured by waves during marine transgression to form a boulder lag that is enclosed in transgressive marine claystone. All four of these examples show that oil was reaching the Earth’s surface during Paddy time where it infiltrated porous sands to a depth of several decimetres, subsequently degrading and oxidizing to form a tough ‘asphalt pavement’ that attracted a burrowing firmground fauna and was eroded into cohesive intraclasts weighing many tens of kilograms. Pyrolysis proved the bitumen cement to be degraded oil, but it was not possible to identify the source-rock using gas chromatography-mass spectrometry. The Paddy ‘asphalt pavements’ are comparable to cements that form in beach sediments in the wake of catastrophic spills from tanker groundings. A direct analogue is provided by Eocene estuarine sediments in Dorset, U.K. There, bitumen-cemented estuarine sands also formed firmgrounds and intraclasts on channel-floor and marine transgressive surfaces.

Abstract The Upper Albian to Lower Cenomanian Hasler and Cruiser shales and intervening Goodrich Sandstone are up to about 800 m thick and constitute most of the upper Fort St. John Group in the foredeep of NE British Columbia. These rocks exhibit pronounced lateral changes in lithology and thickness from west to east, making their lithostratigraphic boundaries highly diachronous. An allostratigraphic approach to correlation, based primarily on discontinuity surfaces represented in wireline well logs, allows subdivision of these broad lithostratigraphic units, and permits correlation of genetic stratal packages across facies transitions, and from subsurface to outcrop. Upper Fort St. John strata spanning the foredeep in British Columbia are shown to be correlative with Viking allomember VD, and with the overlying Westgate and Fish Scales alloformations, previously defined in Alberta on the basis of outcrop, core and wireline log data. Bounding surfaces have also been correlated northward to connect with successions described from the Sikanni and Liard river areas of British Columbia. In the British Columbia study area, pyritic, largely unbioturbated mudstone of allomember VD thickens from 20 to 200 m, indicating syn-depositional flexural subsidence, greatest in the north-western part of the study area. The Westgate alloformation is also dominated by weakly bioturbated mudstone, but includes more common, thin, fine-grained sandstone beds, many with wave- and combined-flow ripples, organized in parasequences up to approximately 20 m thick. Westgate allomembers WA through WD take on a progressively more tabular geometry upward, reflecting diminishing flexural subsidence with time. Allomember WD includes shoreface/delta front sandstones (‘Goodrich Formation’) that extend eastward from the Foothills outcrop for up to 70 km. The Fish Scales alloformation, the base of which is defined by surface FE1, marks an abrupt introduction of sand across a muddy sea floor in response to sea-level fall. The upper boundary of the Fish Scales is the condensed section/downlap surface ‘Fish Scales Upper’ (FSU) beneath the Dunvegan alloformation. The Fish Scales alloformation is a pronounced wedge, thickening westward from approximately 100 to approximately 400 m. It consists mainly of pyritic, unbioturbated mudstone to claystone with a variable content of fish debris, indicative of dysaerobic to anaerobic bottom-water conditions. Ammonites do, however, indicate oxic surface waters. The Fish Scales alloformation comprises allomembers FA and FB, separated by an erosion surface termed the Base Fish Scales Marker (BFSM) which, throughout the Peace River region, has a veneer of chert pebbles that mark a major sea-level lowstand; pebbles were subsequently reworked by marine transgression. A 15 m thick conglomerate-filled channel (or paleovalley?), exposed in Hasler Creek, directly underlies the BFSM surface and is interpreted to be the deposit of a large river that supplied sediment to the lowstand shoreline. In the far west, close to the Foothills, the Fish Scales alloformation becomes sandy and bioturbated, and includes stacked ‘Goodrich’ shoreface sandstone bodies that prograded a few tens of km eastward. Mapped to the SE, allomember FA pinches out close to the Alberta-British Columbia border, whereas allomember FB persists as a thin and highly radioactive phosphatic mudstone, the base of which comprises the merged FE1 and BFSM erosion surfaces. Despite up to about 800 m of subsidence, sedimentary structures throughout the succession show that the sea floor, which formed a homoclinal ramp, remained above storm wave base, attesting to a very high sedimentation rate and effective offshore dispersal of sediment, primarily by storm-driven combined flows. Bivalves of the genus Posidonioceramus have recently been proposed as a marker for the base of the Cenomanian Stage. These fossils are found in the study area as low as the base of Viking allomember VD, which implies that the Albian-Cenomanian boundary may lie at a level significantly lower than the long-accepted ‘Base of Fish Scales’ erosion surface.

Previous studies of Cretaceous sedimentary rocks have used multi-proxy correlation methods to suggest eustatic change, modulated by the c. 400 kyr long eccentricity rhythm. Although numerous authors have inferred eustatic changes on shorter timescales, none have demonstrated synchronous sea-level changes in separate basins on different plates, thousands of kilometres apart. Our study integrates basin-scale, three-dimensional sequence architecture, molluscan biostratigraphy, and carbon-isotope chemostratigraphy to demonstrate synchronous sea-level changes in upper Turonian to lower Coniacian shallow-marine clastic successions in the Western Canada Foreland Basin, and the Bohemian Cretaceous Basin. Depositional sequences in both basins are plotted in a common time domain using an astronomically calibrated age model, allowing direct comparison. In both basins, at least seven major transgressive events can be shown to be synchronous within the limits of combined biostratigraphic and chemostratigraphic resolution. 'Major' and 'minor' sequences of late Turonian age appear to have been paced, respectively, by the long (c. 400 kyr) and short (c. 100 kyr) eccentricity cycles. In contrast, early Coniacian sequences evidence pacing by the c. 38 kyr obliquity rhythm. Stratal architecture suggests that sequences developed in response to eustatic changes of c. 14-20 m at average rates ranging 0.08 to >1.3 m/kyr. At a time of 'warm greenhouse' climate, sea-level change of this magnitude and timescale may not be explicable entirely as a result of thermal- and aquifer-eustasy, and hence glacio-eustasy may also have been a contributing factor. (C) 2021 Elsevier B.V. All rights reserved.

Numerous crocodylian swim tracks occur in the late Early Cretaceous and earliest Late Cretaceous of western North America, in coastal plain deposits marginal to the Cretaceous interior seaways. New as-semblages from the lower Albian Gates Formation, within the Tumbler Ridge UNESCO Global Geopark Quintette site (TRUGG-Q) in British Columbia, Canada, resemble those from Albian-Cenomanian Dakota Sandstone sites in Colorado, Kansas and Utah, and facilitate comparative analysis of how such assem-blages are registered and preserved. The Gates Formation track assemblage formed on the southwest margin of the 'Moosebar Sea' that filled the foredeep of the Western Canada Foreland Basin in the Early Albian. Northeast-prograding sandy deltas were backed by low-lying and heavily-vegetated coastal plains, inhabited by track-making dinosaurs, birds, and crocodylians. The base of a tidal channel-fill at the TRUGG-Q site preserves crocodylian swim traces indicating progression largely parallel to the channel axis, as also inferred for some USA sites. Swim track lengths can be categorized on a shortelong spectrum which reflects swimming behavior, with short, wide toe tip traces indicating firm digging-in of toes ('punting' or 'poling' behavior), and elongate scrape marks indicating light touch-down (or substrate 'raking') by buoyant animals making intermittent contact with a subaqueous substrate. The body lengths size of crocodylian trackmakers can be calculated from rare complete 'walking' track footprints or from the widths, not lengths, in the case of 'swim' traces. The mean size of the Canadian tracks indicates individuals with body lengths in the range of-1.0 m, with larger trackmakers up to-4.0 m more common at the USA sites. (c) 2021 Elsevier Ltd. All rights reserved.

Abstract After prograding for several hundred kilometres during Middle Cenomanian time, the Dunvegan delta complex in north‐west Alberta and adjacent British Columbia experienced stepwise transgression, commencing at about the Middle to Late Cenomanian boundary. Progressive drowning of the delta complex is recorded by Dunvegan allomembers B and A, each comprised of three simple depositional sequences, bounded by composite subaerial unconformity/flooding surfaces. Each sequence represents an array of deltaic depositional environments. Delta‐front sandstones preserve little evidence, such as hummocky cross‐stratification, for powerful storm wave action, although wave and combined‐flow ripples are common. Delta‐front sandstone bodies tend to be smaller and lobate in the lower part of the studied interval, and larger and more linear near the top. This suggests increasingly effective wave‐driven redistribution of sand as more open‐marine conditions were gradually established. The top surfaces of allomembers B and A are locally incised by sandstone‐filled palaeovalleys up to 19 m deep; river incision may have been a response to relative sea‐level fall and/or a change in the ratio of discharge to sediment load. Overall, the shoreline described a broad arc, open to the south east, with the sense of shoreline migration north‐west to south‐east. For each sequence, the shoreline migrated an average of 80 km between transgressive and regressive limits. The transgressive limit shows a progressive landward offset of about 15 km per sequence, culminating in complete drowning of the delta system above sequence A3. Isopach maps show that syn‐depositional tectonic subsidence rotated the basin down to the south‐west; palaeogeographic maps show, however, that the sea floor sloped to the south‐east, implying that sediment redistribution effectively filled all tectonically generated accommodation and maintained a south‐east inclined depositional surface. Transgressions and regressions across this surface were therefore driven primarily by eustasy rather than pulses of tectonic subsidence. Simple calculations based on inferred alluvial gradients of 10–20 cm/km suggest that eustatic excursions of ca 8–16 m would have been sufficient to generate sequence thicknesses on the order of 10 m. Limited geochronologic and biostratigraphic control suggests that the six simple sequences that form Dunvegan allomembers B and A each represent an average of about 41 kyr, suggesting that the orbital obliquity cycle was the primary control on high‐frequency sea‐level cycles.

An ephemeral, ice-based flume developed in Medway Creek (London, Ontario) during a February thaw when water at ~1oC flowed over the ice surface forming a < 10 cm deep, ~ 3 m wide channel. Eroded muddy bank sediment, composed of silt to medium-sand sized aggregates, formed linear streamers that revealed streaks in the boundary layer. In water 6-8 cm deep with a flow velocity of ~ 8-12 cm/s, mud aggregates were molded into lunate, transverse, and ovoid ripples a few mm high. Clear water allowed mud aggregates in streamers to be observed accreting to, and migrating over mud ripples. Downstream of larger ripples, mud streamers were swept clear of the bed, perhaps due to vortices shed by the ripple. Where flow exceeded ~ 12 cm/s, mud ripples were gradually replaced by mud aggregate streamers which in turn were washed out in an area of faster (undetermined rate) flow. The flow conditions and bedforms in this ephemeral, natural flume are closely comparable to those described from laboratory flumes at 25oC; however the increased viscosity of water at 1oC may alter the stability field of mud ripples.

The late Cenomanian Kaskapau Formation records transgression of the Greenhorn Sea over deltaic strata of the Dunvegan Formation. However, stratigraphic and paleogeographic details of the initial stages of this profound reorganization of basin geography have not previously been determined. The basal Kaskapau strata above the Dunvegan Formation are assigned to the informal allostratigraphic 'A-X unit'. Subsurface and outcrop correlation shows that five regionally-mappable allomembers, bounded by marine flooding surfaces, can be mapped within the A-X unit which, as a whole, forms a wedge that thickens south-westward from < 5 to 60 m over approximately 300 km. Within each allomember, three main depositional environments can be distinguished. Sandy heterolithic facies in the north and west form few-metre scale, upward-shoaling successions that contain abundant brackish-water molluscs and are capped by paleosols and dinosaur-trampled surfaces. These rocks represent river-dominated deltas that prograded into a low-energy embayment, about 200 x 200 km, and open to the SE. The central part of the embayment accumulated mud-dominated heterolithic successions with a restricted fauna of lingulid brachiopods and inoceramids, suggestive of turbid, low-energy and reduced salinity conditions. The muddy facies enclose isolated units of well-sorted fine- to very fine-grained sandstone that form NE-SW elongate bodies up to 170 km long, 50 km wide and 11 m thick. Each sandstone body is interpreted to represent the shoreface of one or more strandplains and/or wave-dominated deltas that developed across the mouth of the embayment at sea-level lowstand. Isopach mapping shows that the A-X unit, as a whole, forms a prismatic wedge, thickest adjacent to the fold and thrust belt in the SW. This geometry indicates that accommodation was created by spatially uniform flexural subsidence in response to a linear tectonic load striking NW-SE. Isopach maps of individual allomembers, however, show that subsidence took place in a more complex pattern, with localized depocentres of approximately 100 km radius that suggest brief periods of subsidence before being superseded by a new depocentre along-strike. This pattern may reflect temporally and spatially discontinuous deformation in the adjacent orogenic wedge. During each relative sea-level cycle, the shoreline oscillated by about 150-200 km in a NW-SE direction. Shoreline movement was perpendicular to the direction that would be expected (i.e. SW-NE), if flexural subsidence had been the dominant control on relative sea-level change. This geometric relationship suggests that allomembers were generated by high-frequency, high rate eustatic changes that were superimposed on a lower rate of flexural subsidence. Isopach maps spanning Dunvegan allomembers C, A+ B, and the Kaskapau A-X unit show a near 90 degrees anticlockwise rotation of isopleths. This abrupt re-orientation of the axis of flexure marks a new phase of subsidence linked to the onset of NE-directed thrusting that was driven by dextral transpression across a restraining bend in the Northern Rocky Mountain Trench fault. This new stress regime was in turn linked to a change in the convergence direction between the North American and Farallon plates.

The great Late Cenomanian transgression of the Greenhorn Sea is recorded in western Alberta and adjacacent British Columbia by shallow-marine deposits of the lower Kaskapau Formation that blanket underlying deltaic deposits of the Dunvegan Formation. Overlying the basal 'A -X unit' of the Kaskapau are the allostratigraphically-defined Doe Creek and Pouce Coupe units, each of which forms a SW-thickening wedge up to about 100 m thick. The Doe Creek and Pouce Coupe units comprise, respectively, 10 and 8 allomembers, each defined by a marine flooding surface. Doe Creek facies are organized in upward-shoaling successions less than about 10 m thick. Shallow-water marginal-marine facies prevail in the west and north, with transgressive mudstones overlain by thin, sharp-based shoreface sandstones that, in places, contain roots and dinosaur tracks. In some instances, marine mudstone grades directly upward into a rooted, muddy paleosol. Mapped regionally, these facies define a broad, shallow, SE-facing embayment approximately 200 x 200 km. 'Trace and molluscan fossils suggest that salinity was below normal marine level, and benthic fauna were, at times, stressed by high turbidity and sedimentation rate. The central part of the embayment is dominated by offshore mudstone that encloses isolated, sharp-based lenticular bodies of clean sandstone up to 9 m thick. Sandstones are ovoid to elongate in plan view, and commonly extend many tens of km N-S or NE-SW. These sandstones are interpreted as top-truncated lowstand delta and strandplain deposits that became isolated on the outer ramp following marine transgression. The sharp base of most shoreface sandstones implies that deposition took place during relative sea-level fall. All Doe Creek allomembers thicken into the foredeep, with the tectonically-flexed surface inclined to the SW. Palaeogeographic maps show, however, that lowstand shorelines were oriented approximately NE-SW, and that the depositional surface sloped to the SE. This geometric relationship implies that, despite tectonic subsidence increasing to the SW, the rate of sediment supply and efficiency of redistribution were able to maintain a wave-graded surface sloping to the SE. Because shorelines migrated across, rather than up and down tectonic dip, allomembers are interpreted to have been controlled primarily by high-rate, high-frequency eustatic cycles, rather than by tectonic movements. The Pouce Coupe unit represents a complete reversal of basin paleogeography relative to the Doe Creek. Pouce Coupe shoreface sandstones are stacked in a linear belt just west of 120 degrees W and grade westward into offshore muddy facies that thicken to the WSW. The entire unit is truncated eastward by the K1 unconformity and is absent over most of Alberta. It is postulated that the Proterozoic Kiskatinaw domain acted as a crustal weak zone, forming a hinge that defined the eastern margin of the Pouce Coupe flexural depocentre.

Across southern Alberta and Saskatchewan, rocks of Late Albian age (including the Joli Fou, Viking, Bow Island and Mill Creek formations) represent depositional environments ranging from alluvial to deltaicnearshore to offshore marine. Because of radical facies changes, it is difficult to establish age relationships amongst these various lithostratigraphic formations. Building upon an existing allostratigraphic framework developed by Boreen and Walker (1991), Roca et al. (2008), Buckley and Plint (2013), and Vannelli (2016) have modified the original allostratigraphic scheme and extended correlation of Upper Albian strata to northern Alberta and NE British Columbia. The investigation reported here builds upon these allostratigraphic studies and extends across 56,000 km of south-central Alberta and Saskatchewan. Preliminary results show that early Late Albian Joli Fou sediments form a subtly SEthickening wedge whereas lower Viking allomembers VA and VB are approximately sheet-like, with local thickening to the SW reflecting local deltaic depocentres. Viking allomember VD contasts markedly with allomembers VA and VB because it forms a prominent, elongate SW-thickening wedge that extends across Alberta into Saskatchewan, within which, parasequences onlap towards the NE. The NE limit of the wedge is abrupt, defined by a prominent lineament, or hinge-line. To the south of the hinge, overlapping sandy deltaic lobes can be mapped whereas to the north of the hinge, the rocks are mudstone-dominated.