The evolution of catchment- depositional system relationships on the dip slopes of intra- rift basement highs: An example from the Frøya High, Mid- Norwegian rifted margin

Basement highs form one of many potential sediment source areas during the evolution of continental rifts and rifted margins and add to the topographic complexity typical of active rifts. Footwall basement highs acting as a source area to sedimentary systems in the hangingwall of major faults have been documented in many systems worldwide. However, the back- tilted footwall dip slopes of such highs have received comparatively little attention. Here, we investigate a subsurface case study from the Norwegian continental shelf, where catchments and shallow marine syn- rift sedimentary systems on a dip slope are preserved due to early transgression of an intra- rift high. At the onset of Late Jurassic rifting, the Frøya High emerged as a prominent, N- S trending, 25 km- wide basement high tilted towards the east in response to several kilometres of displacement along the Klakk Fault


| INTRODUCTION
The evolution of rift basins promotes a wide variety of sediment routing and depositional systems, which can vary considerably over relatively short temporal and spatial scales (e.g. Barrett et al., 2019;Gawthorpe et al., 1994;Gawthorpe & Leeder, 2000;Nøttvedt et al., 2000;Ravnås & Steel, 1998). High rates and magnitudes of subsidence in the hangingwall of active structures produce the greatest areas of accommodation, with many studies focussing on the delivery of sediment from drainage catchments on nearby footwall crests (e.g. Barrett et al., 2021;Bilal et al., 2020). However, the uplift and rotation of the backtilted footwall dip slope also offer accommodation for sedimentary systems fed by drainage directed away from the footwall crest. (Gawthorpe & Leeder, 2000;Muravchik et al., 2018;Rapozo et al., 2021;Ravnås & Steel, 1998;Smyrak-Sikora et al., 2018, 2021. The tectono-sedimentary setting of dip slopes developed along the footwall of major faults is considerably different to that of immediate hangingwall systems such as fault-scarp degradation-related fans or rift-margin deltas. Lower gradients and subsidence rates, of back-tilted slopes in the footwall of major normal faults, lead to a greatly enhanced sensitivity to eustatic changes across broader, shallow landscapes compared with hangingwall depocentres and commonly host shoreline depositional systems (Bell et al., 2017;Fernández-Blanco et al., 2020;Gawthorpe et al., 1994;Smyrak-Sikora et al., 2021). However, the controls upon the spatial variability in the depositional environment and resultant stratigraphic architecture of dip slope shoreline systems in rift settings remains comparatively unclear, especially those flanking large intra-rift basement highs Nøttvedt et al., 2000). With the exception of few localised outcrop (Jackson et al., 2005;Muravchik et al., 2018;Smyrak-Sikora et al., 2021) and subsurface examples (Chiarella et al., 2020;Nøttvedt et al., 2000;Ravnås & Steel, 1998), there are few studies, which address the scales, along-strike variability and potential controls on such syn-rift depositional systems.
Offshore Mid-Norway, at the necking domain of the Norwegian rifted margin, the Frøya High hosts a number of fringing syn-rift depositional systems, formed on the footwall dip slope of a major active rift margin bound by the Klakk and Vingleia fault complexes during the Late Jurassic and Early Cretaceous (Figure 1; Blystad et al., 1995;Bell et al., 2014;Elliott et al., 2017;Muñoz-Barrera et al., 2020;Jones et al., 2021). Except for stratigraphy at the northern end of the Frøya High, near the Draugen ridge (Chiarella et al., 2020), syn-rift sediment routing and depositional systems on the eastern flank of the Frøya High have received little attention. The eastern flank of the Frøya High allows for a coherent investigation into the link between sediment source area and down-dip deposition, partly due to the partial preservation of drainage catchments on the Frøya High following a prolonged period of flooding and burial of the Frøya High at the end of the Jurassic (Jones et al., 2021). Excellent 3D seismic data coverage and recent exploration drilling in 2022 have allowed for re-evaluation of stratigraphic relationships between the Frøya High and basinal areas with improved constraints through newly acquired biostratigraphic, wireline and sidewall core samples within the Upper Jurassic, Viking Group stratigraphy. Here, we integrate this seismic and well data to investigate syn-rift sedimentation along a of the protracted backstepping observed in other dip slope systems. We postulate that different structural configurations of dip slope systems, being footwall uplift or hangingwall subsidence driven, may drive the strongly progradational character of the deltaic systems on the Frøya High. The Frøya High example highlights the need to constrain primary sediment input points to aid the interpretation of volumetrically significant, but short-lived and subtle depositional systems, especially within complex, tectonically active settings. EAGE HENSTRA et al. footwall dip slope of a major rift-margin normal fault, the Klakk Fault Complex (Figure 1). Biostratigraphically constrained deposits and surfaces permit the mapping and characterisation of three-dimensional variability of syn--rift drainage catchments in the footwall of the Klakk Fault Complex and dip slope the associated dip slope depositional systems at the southern and central part of the Frøya High. Finally, integration and comparison of observations from the Frøya High with other dip slope systems from other rift basins demonstrate the variability and role of structural setting in determining the architecture of dip slope sedimentary systems.
East of the Frøya High lies the southern tip of the 50 km-wide Froan Basin that extends for 250 km north along the Trøndelag Platform (Blystad et al., 1995). The Froan Basin is underlain by thick continental crust and contains Palaeozoic to Cenozoic sedimentary successions. Unlike the western, fault-controlled margin of the Frøya High, the south-eastern margin of the Frøya High passes into the Froan Basin across a gently east-dipping basementsediment contact cut by several minor, small-displacement normal faults (Figure 1). In the northern part of the Froan Basin, this boundary becomes more distinct where the Froan Basin is delimited by a narrow horst between the Vingleia Fault Complex (VFC) and Froan Fault (Figure 1; Bunkholt et al., 2022;Elliott et al., 2017;Gernigon et al., 2020;Osmundsen et al., 2021;Wilson et al., 2015).
The Mid-Norwegian margin evolved through several rift episodes following the Devonian collapse of the Caledonian mountain range (Bell et al., 2014;Coward et al., 2003). A major rift episode occurred during the Late Permian-Early Triassic leading to crustal stretching and the formation of NE-SW-trending rotated fault blocks on the mid-Norwegian margin (Brekke, 2000;Brekke & Riis, 1987;Coward et al., 2003;Halland et al., 2014;Peron-Pinvidic et al., 2013). During this time, a smaller basin was generated on the Frøya High informally known as the Almond Basin, a 2-5 km-wide and 0.5 s TWT thick half graben filled largely with, unconfirmed but likely, Permian to Middle Jurassic stratigraphy ( Figure 1; Blystad et al., 1995). Following a period of relative tectonic quiescence from the Mid-Triassic through to the Early Jurassic, extension resumed within the Middle Jurassic through to the Late Jurassic (Brekke, 2000;Coward et al., 2003;Faerseth, 1996;Roberts et al., 2009). During this Middle-Late Jurassic phase, the main tectonic elements such as the Møre Basin, the Frøya High and the Halten Terrace started to form as the rift reached a widely documented 'thinning' phase and deformation became localised (Bunkholt et al., 2022;Muñoz-Barrera et al., 2020;Osmundsen & Péron-Pinvidic, 2018;Osmundsen et al., 2002;Peron-Pinvidic et al., 2013). The Klakk, Vingleia and Bremmstein fault complexes accumulated substantial displacement at this time and became the location of crustal necking leading to the uplift of the Frøya High that emerged as a prominent high, during the latter part of the Middle and throughout the Late Jurassic-Early Cretaceous within a broader seaway encompassing the approximately 250 kmwide and 1500-km-long Trøndelag Platform separating Norway and Greenland (Figure 1; Bunkholt et al., 2022;Chiarella et al., 2020;Coward et al., 2003;Jones et al., 2021;Nøttvedt et al., 2008;Osmundsen & Ebbing, 2008;Peron-Pinvidic et al., 2013).
The Froan Basin itself was only moderately tectonically active during the Late Jurassic-Early Cretaceous rift episode, with a minor component of uplift on its western margin onto the Frøya High in response to uplift in the footwall of the KFC (Blystad et al., 1995). Following the Late Jurassic-Early Cretaceous rift episode, deformation became focused on the distal domain of the Norwegian Margin, west of the Frøya High (Brekke & Riis, 1987;Brekke et al., 1999;Blystad et al., 1995;Faerseth, 2020;Osmundsen et al., 2002).

Jurassic-Early cretaceous rift episode: The Viking Group
Depositional products of the sedimentary systems that were active on the eastern margin of the Frøya High during the Jurassic-Early Cretaceous rift episode are part of the Middle-Upper Jurassic Viking Group ( Figure 2). In the study area the Viking Group either directly overlies the Middle Jurassic Fangst Group, Triassic sediments known as 'Red' and 'Grey' Beds in local stratigraphic terminology (Blystad et al., 1995) or granitic basement (Figure 2b). The oldest formation of the Viking group is the Callovian-Oxfordian Melke Formation that is commonly interpreted to record the transition from prerift to rift initiation in the area of the Frøya High (Blystad et al., 1995;Chiarella et al., 2020;Corfield et al., 2001;Jones et al., 2021). The Melke Formation, where present, is overlain by the Spekk or Rogn formations, which have been linked to 'rift climax' and late rift phases (Jones et al., 2021). The Spekk Formation is largely mud-prone, high in organic content and is widespread on and around the Frøya High (Jones et al., 2021;Muñoz-Barrera et al., 2020, Bunkholt et al., 2022. The Rogn Formation is a distinct sand unit that commonly sits stratigraphically within the Spekk Formation on the eastern flank of the Frøya High (Figure 2b,c, Chiarella et al., 2020). The Rogn Formation is typically interpreted to represent Late Jurassic-Early Cretaceous deposits of shoreface to offshore-bar environments (Chiarella et al., 2020;Jones et al., 2021) in a narrow seaway between footwall block islands and the Late Jurassic coastline of the Norwegian mainland (Bunkholt et al., 2022;Chiarella et al., 2020;Faerseth, 2021). The Viking Group is capped by a regional unconformity, commonly referred to as the Base Cretaceous Unconformity (BCU), which is overlain by deposits of the mud-prone Lyr and Lange formations of the Cromer Knoll Group. In most parts of the study area, the Cromer Knoll Group onlaps the BCU. However, the BCU can have a conformable appearance picked out by a high amplitude negative reflector (Figure 2a).

| Seismic and well data
This study makes use of a reprocessed 3D seismic reflection survey that covers most of the southern part of the Frøya High from the KFC in the west to the Froan Basin in the east (Figure 1a). Seismic data are in normal polarity, displayed in SEG normal convention (downward increase in acoustic impedance = positive amplitude). The wells 6306/6-1 and 6306/9-1 provide tie-points for the key seismic markers: Top basement, Base Viking Group, Top Viking Group (BCU) and base Upper Cretaceous and Top Cromer Knoll Group ( Figure 2a; Table 1).
Well 6306/6-1 is located on the eastern Frøya High and is used to identify the seismic expression of key stratigraphic events within the syn-rift interval (Figures 1b  and 2). The sonic and density logs of well 6306/6-1 provide a record of acoustic impedance contrasts within the stratigraphic interval of interest and generation of a synthetic seismogram that allows the characteristics of the key stratal surfaces to be constrained ( Figure 2 and Table 1).

| Structural restoration
Two-way time structure maps ( Figure 3) generated by detailed manual interpretation of 3D seismic reflection data were depth converted using a velocity model that is based on stacking velocities (Ashcroft, 2011;Johnson & Hansen, 1987;Marsden, 1989). The accuracy of the resultant depth-converted surfaces is independently verified by wells 6306/6-1 and 6306/9-1: the difference between the depth-converted seismic horizons and their actual depth in the wells is less than 20 m. We apply a crude structural restoration through a simple 'rigid body' rotation of a depth-converted Top Viking Group surface is applied to remove the westward, post-rift tilting associated with continued burial and post-rift thermal subsidence throughout the latter part of the Cenozoic (Bell et al., 2014;Brekke, 2000). This provides an approximation of the topography as it existed during the Late Jurassic. Deformation effects related to compaction and flexure during burial and post-rift thermal subsidence are expected to be minimal and consistent across the study since the area of interest (15 × 20 km) is substantially smaller than the flexure of the Norwegian margin during post-rift subsidence (Bell et al., 2014;Brekke, 2000).

| SEISMIC MAPPING
The seismic-to-well tie generated for 6306/6-1 ( Figure 2) is used to characterise the seismic expression of three key stratigraphic contacts: Basement, base Viking Group and top Viking Group/'BCU'. All three reflectors are mapped with high confidence in the eastern part of the seismic reflection survey (Figure 3). F I G U R E 2 (a) Well-to-seismic tie for the stratigraphic interval of interest. A synthetic seismogram is generated from the sonic and density logs of well 6306/6-1 (location in Figures 1 and 3). Key surfaces that are recognised in the well are thus linked to their corresponding reflector in the seismic reflection survey shown in (b). (c) Stratigraphic chart highlighting the lithostratigraphic nomenclature for the Norwegian Margin in the Southern Norwegian Sea (modified from Gradstein et al., 2010 andGradstein, 2017). TVD, true vertical depth; TWT, two-way time; FH, Frøya High; SR, Sklinna Ridge; GH, Gossa High. Tectonic events summarised by Bunkholt et al. (2022).

| Top basement TWT structure map
The contact between the crystalline basement and sedimentary cover corresponds to a positive impedance contrast and a high amplitude, peak (red) reflection where the basement is overlain directly by Upper Jurassic/Cretaceous strata (Table 1, Figure 2). Where it is overlain by older Mesozoic strata the contact generates a reflection with lower amplitude (Figures 2 and 3). The seismic expression of the top Basement reflector can therefore be used to help distinguish whether the basement is overlain by Upper Jurassic/ Cretaceous strata or older Mesozoic strata (Figures 2 and 3). In the eastern part of the study area, the top Basement surface underlies the Late Palaeozoic-Mesozoic infill of the Froan Basin (Figures 1c and 4). There, the contact between the basement and sedimentary cover is a nonconformity, relatively smooth and dips to the east (e.g. Figures 1c and  4). In the western part of the study area the basement surface is part of the footwall scarp of the KFC generating the Frøya High ( Figure 3a). Here, the basement surface dips to the west, is strongly undulating and is onlapped by Upper Jurassic (Viking Group) and Lower Cretaceous strata ( Figure 1c). Within the central part of Frøya High, the basement surface dips to the east within a half graben, which is informally known as the 'Almond Basin' and is overlain by pre-Upper Jurassic sedimentary strata.

| Base Viking Group TWT structure map
The base Viking Group surface commonly represents a positive impedance contrast and is expressed mostly as a peak (red) event, but its seismic expression is variable, and it may be expressed as a trough (blue) event locally ( Figure 2). This variable expression is related to the unconformable nature of this stratigraphic contact, juxtaposing different lithologies across the contact.
The base Viking Group surface is relatively smooth and dips to the northwest across much of the Frøya High and margin of the Froan Basin where the Viking Group overlies the Middle Jurassic conformably. However, with greater proximity to the Frøya High the base Viking Group is increasingly recognised as an angular unconformity (Figures 1c, 2 and 4). Biostratigraphy from 6306/6-1 and 6306/9-1 records a hiatus across this surface that may span the Bajocian (Statoil, 1994. In areas along the southern part of the Frøya High, and underlying an N-S-oriented concave-up trench on the western edge of the Froan Basin, the Base Cretaceous Unconformity ('BCU') has eroded down to the pre-Upper Jurassic interval. As a result, the Base Viking Group surface is coincident with the Top Basement surface where Middle Jurassic stratigraphy is absent ('deeply eroded' areas in Figure 3b).
The distribution of the Viking Group is more complex on the Frøya High than in the Froan Basin. Where the Viking Group is preserved it may rest either on the crystalline basement or on Lower Mesozoic stratigraphy (e.g. Figure 4). The latter form is most common, especially overlying the Almond Basin where the greatest areal extent of Viking Group stratigraphy is preserved on the Frøya High ( Figure 5). Where the Viking Group overlies the Permian and Middle Jurassic of the Almond Basin, the Base Viking Group surface is an angular unconformity of around 15 degrees with older stratigraphy within T A B L E 1 Amplitude characteristics of the key stratigraphic horizons in 6306/6-1 and surfaces used in this study.

Well top name
Amplitude at 6306/6-1 Character Variability Zero-crossing to high positive amplitude peak from low-amplitude noisy data Strong lateral continuity Laterally consistent the Almond Basin ( Figure 4). Upper Jurassic, Viking Group stratigraphy noticeably onlaps relative palaeohighs on the central and western part of the Frøya High and three small (1-2.5 km wide) highs on the eastern part of the Frøya High on the flank of the Froan Basin ( Figure 3b).

| Top Viking Group (BCU) TWT structure map
The top Viking Group/'BCU' surface marks a negative impedance contrast and corresponds to a trough (blue) event at the intersection with 6306/6-1. However, its amplitude changes spatially recording the variable lithology of underlying and overlying stratigraphy ( Figure 2). It represents an erosional surface that is generally broadly parallel to the underlying strata across much of the Frøya High and margin of the Froan Basin, but locally is a down-cutting surface, eroding deeply into the substrate, that has a concave-up geometry with relief of 50-200 m in parts of the southern Frøya High and within the Froan Basin (X in Figures 3c and 4). Overall, the surface deepens to the west with localised regions of the steepening gradient along the eastern edge of the Almond Basin and along an N-S running ridge 2-3 km west of the western edge of the Almond Basin (D in Figure 3c). There are several regions of onlap onto relative palaeohighs, typically 1-5 km in diameter and ~40-50 m high on the eastern side of the Frøya High, and a small cone-shaped, 2-km-wide region west of the Almond Basin across an area of the steepening gradient of the BCU (Figure 3c).

| The Viking Group isopach map
The Viking Group isopach map reveals substantial spatial variability in thickness ( Figure 3a) with a relatively large, 20 km wide, depocentre overlying and east of the previous Almond Basin (Depocentre O; Figure 5a), with ca. 150 m of Viking Group stratigraphy. In the southern part of the Almond Basin, these strata are truncated and deeply eroded by the BCU forming a ~15 km-long, 10 km-wide depression that cuts down progressively deeper towards the southwest reducing the thickness of the Viking Group to <10 m and locally making it absent (Figures 3b,c-5a). A few kilometres to the east of the Almond Basin are three accumulations of Viking Group strata up to 300-250 m thick each measuring ca. 5 km in width (Depocentres A, B and C; Figure 5a). These accumulations show approximately fan-like geometries in plan view, with radial thinning from their centres, and are separated by narrow (1-3 km) wide regions of zero or very limited thickness (Areas I, II and III Figure 5a). Immediately east of these wedges' angular truncations of Upper Jurassic (Viking Group) and Middle Jurassic stratigraphy and a substantial decrease in Viking Group thickness suggest widespread erosion along an N-S trending depression (Figures 4b and 5). The N-S depression (Area E; Figure 4b) has a jagged western edge, approximately 25 km long and 3-5 km wide with a maximum relative depth of ~500 m (Area E; Figure 5b). Farther to the northeast, closer to the centre of the Froan Basin, the Viking Group is preserved and exhibits a relatively uniform thickness between 30-40 m over a 5 × 5 km area (Depocentre D; Figure 5a). Overall, the variable thickness patterns are interpreted to represent both nonuniform deposition in Jurassic times and subsequent erosion during the Early Cretaceous.

| The Cromer Knoll Group isopach map
The top of the Viking Group, commonly referred to as the BCU, marks the base of the Cromer Knoll Group isopach. The BCU is marked by onlap by the lowest part of the Cromer Knoll Group in parts of the crestal areas of the Frøya High (e.g. D in Figure 4c) and is conformable within areas on the eastern margin of the Frøya High and in the Froan Basin (e.g. F in Figure 4). The Cromer Knoll Group is relatively thin (25 m or less) or absent over the crestal and eastern part of the Frøya High (Area A, Figure 4). Conversely, in the Froan Basin to the east, there are substantially thicker accumulations of Cromer Knoll Group stratigraphy (Area E and F, Figures 4 and 5b). The large, N-S trending erosional feature hosts up to 500 m of Cromer Knoll stratigraphy, which onlaps the BCU within Area E, becoming generally thicker towards the south of and east of the feature (Figures 4 and 5b). To the northeast within the Froan Basin, thicknesses of up to 300 m of Cromer Knoll group are observed (Area F in Figure 5b). This spatial pattern of the thickness of the Cromer Knoll Group and onlap is interpreted to reflect the progressive flooding of the Frøya High, with the Frøya-Froan margin and the crestal part of the Frøya High being one of the last areas to be flooded during the early Cretaceous (Bell et al., 2014;Brekke, 2000;Jones et al., 2021). Unlike the Viking Group isopach map, the lateral variability in the thickness of this interval is attributed mostly to nonuniform deposition on a variable underlying topography with the majority of the Lower Cretaceous deposits of the Cromer F I G U R E 5 Isopach maps of (a) Late Jurassic, Viking Group, generated by depth converting and combining the base and top Viking Group TWT structure maps shown in Figure 3b,c, respectively. (b) Early Cretaceous, Cromer Knoll Group, generated by depth converting and combining the top Viking Group TWT structure map ( Figure 3c) and an autotracked TWT surface of the base Shetland Group reflector. Key depocenters referred to in the text are labelled A, B, C, O-G.
Knoll Group being nonerosional and internally conformable for the most part. The Froan Basin in the easternmost part of the study area continued to receive sediment since earliest Cretaceous times (Depocentre F; Figure 5b; Skarbø et al., 1988). The large, N-S-trending erosional feature is interpreted to have been excavated during the early stages of the Early Cretaceous by the removal of Upper Jurassic, and older strata had become filled by the end of the Early Cretaceous as evidenced by the onlap of Cromer Knoll reflections onto the BCU on either side of the erosional depression (Depocentre E; Figures 4 and 5b).

| Observations
The seismic stratigraphic architecture within the Viking Group fan-shaped, wedges (A, B and C; Figure 5a) is relatively complex in comparison to the expression of the Viking Group elsewhere ( Figure 6). By flattening the seismic on the Top Cretaceous horizon (Top Shetland Group) we can approximately restore the orientation of the reflectors that make up the wedge to their original geometry before the area became tilted towards the west (Figure 7). On the flattened data, the western and upper part of the wedges consists of parallel, horizontal to sub-horizontal reflections that onlap the basement/ base Viking Group surface to the west. To the east, the wedges consist of east-dipping, inclined, sigmoidal reflections that downlap the basement/base Viking Group (Figures 4, 6 and 7). The maximum (nondecompacted) thickness of the wedges amounts to ~250 m (Figures 4  and 7a) and the internal sigmoidal reflectors dip ca. 10°. Heights of inclined reflectors vary from ~30-200 m. When viewed in aNE-SW section u, reflections have apparent dips in opposing directions, away from the approximate centre of each wedge, and generate broadly convex-up internal morphology with internal complexity and surfaces marked by downlaps ( Figure 8).
Well 6306/9-1 penetrates Wedge B and recovers approximately 155 m of Middle to Upper Jurassic, which lies directly on the crystalline basement ( Figure 9). Biostratigraphic and sedimentological analysis of sidewall cores is integrated with wireline and borehole image data into a summary log in Figure 9. The Middle Jurassic stratigraphy comprises a sub-horizontally bedded basal section following bed restoration, from ~941 to 967 m depth, of interbedded fine-medium-grained sandstone and siltstone rich in carbonaceous material and root traces, which is capped by a coarser grained, upward fining gravelly sandstone rich in carbonaceous material but also calcareous shell fragments (Figure 9). Palynological observations in this section show a substantial presence of brackish water indicators typical of a marginal marine assemblage and are dated as Callovian. The basal section transitions abruptly at ~940 m depth into a clay-rich, nonbioturbated siltstone with low palynological diversity at the onset of a largely silt and clay-prone, ~45 m thick Oxfordian section becoming increasingly bioturbated and sand-prone upwards (Figure 9). This is overlain by an abrupt transition at 895 m into sand-prone sidewall core samples and consistently low gamma-ray values with a marked reversal in neutron-density separation (Figure 9). This section is dated as Kimmeridgian-Volgian with the extensive reworking of Oxfordian palynomorphs and coincides with the position of well-recognised eastward dipping, downlapping reflections in the seismic data ( Figure 6). In the well, this is recognised as a change of restored bedding dips to steeper (c. 25°), exclusively eastward dips. The Kimmeridgian section can be split into three sub-units (Figure 9) comprising medium-very F I G U R E 9 Summary well log for the Late Jurassic stratigraphy of 6306/9-1 compiled from wireline, sidewall core, biostratigraphic and borehole imaging observations. FMI, formation microresistivity image; UBI, ultrasonic borehole imager. coarse sandstone-prone packages separated by the shallow dipping (c. 10°) 1-2 m thick argillaceous sandstone beds. The three sub-units broadly demonstrate a steepening to shallowing restored bedding trend punctuated by shallow restored bedding within the intervening argillaceous sandstone intervals (Figure 9). Sandstone beds are typically coarse-grained, granule-bearing and poorly sorted, with occasional development of cross-stratification and abundant organic material. The section returns at 847 m depth to dominantly fine-to medium-grained sandstone that is still granule-bearing and displays planar to cross-stratification coincident with a move into shallowly eastward dipping, planar reflectors in seismic section (Figures 6 and 9). However, this package ultimately becomes Ryazanian-aged and increasingly rich in glauconite pellets and is more intensely bioturbated (Chondrites and Palaeophycus). This is capped at 819 m depth by a relatively thin, 7 m section of intensely bioturbated, coarse to very coarse, sub-horizontally bedded sandstone ( Figure 9). The sandstone itself is overlain by gravel-and pebble-rich sandstones (813-815 m) with broken shell fragments and heavy minerals that account for extremely high gamma readings (Figure 9). This is overlain by a subhorizontally bedded claystone with pervasive Chondrites at 812 m depth, coinciding with the position of the strong negative amplitude associated with the Top of the Viking Group in seismic (Figures 6 and 9).
The Top Viking Group/BCU structure map and isopach maps show that the eastern edge of the wedges has an irregular, jagged, erosional, appearance in the plan view marking the edge of the N-S erosional feature highlighted by the base of Depocentre E (X in Figures 3c and 10). The N-S erosional feature has a connection along its western edge to 0.5-1.5 km wide, smaller scale erosional features trending NW-SE or E-W, which shallow to the west onto the Frøya High. The irregular nature of the BCU at the eastern edge of the wedges also hosts numerous, relatively small (ca. 100 m wide and 500 m long), amphitheatreshaped depressions that shallow westwards ( Figure 10). Isopach maps reveal that the Viking Group is thinner within these features (Figures 5a and 10) and that they are filled in with, younger, Cromer Knoll stratigraphy, that onlaps this surface (Figures 5b and 10b).

| Interpretations
We interpret the Viking Group wedges with their internal sigmoidal reflections as clinoforms. Based on their fan-shaped plan view geometry of the three wedges (A, B and C and D; Figures 5a and 7) and eastward dipping and prograding foresets, the wedges are interpreted as eastward prograding deltaic packages. Observations of carbonaceous, rooted heterolithic sandstones from 6306/9-1 suggest the deltaic clinoforms are preceded by a Callovian-aged coastal plain containing occasional fluvial channels, which became transgressed in the Oxfordian establishing a largely dysoxic, mud-prone shelf and increasingly sand-prone prograding shoreface rich in bioturbation. This backstepping is followed by the strongly progradational deltaic clinoforms observed on seismic (Figures 6 and 7b), which are seen to be that of a coarsegrained sand-prone and poorly sorted, eastward prograding delta initiated during the Kimmeridgian (Figure 9). The observed height and relatively steep (>10°) angle of the clinoform foresets indicate progradation into a waterbody up to ~200 m deep towards the upper, later part of the Viking Group section. Smaller foresets (from 30 to 50 m) during the onset of the clinoform packages ( Figure 6) indicate periods of less accommodation early in the history of the deltas increasing in height through younger clinoforms, with foreset height through the deltas depending on the interaction of sea-level and seafloor topography at a given time (Figures 7 and 9). Argillaceous sandstones intersected in the wells demonstrate similar thicknesses and characteristics to reactivation surfaces separating foreset packages documented in fan deltas (e.g. Backert et al., 2010;Barrett et al., 2019;Gawthorpe et al., 2017). Similarly, the scales (10-30 m thick) and lithology (coarse-grained, gravel-pebble prone sandstones) of the steeper cross-stratified sub-units are comparable to that of Gilbert-type fan deltas with steeply dipping delta fronts comprising a broad range of gravity flow deposits (Barrett et al., 2019;Chiarella et al., 2020;Gobo et al., 2014;Massari & Colella, 1988;Nemec, 1990;Rohais et al., 2008).
The arcuate, amphitheatre-shape of some of the erosional features (Figures 5 and 10) resemble the scars left by slumping of unconsolidated material observed on modern subaqueous depositional slopes (e.g. Biscara et al., 2012;Gales et al., 2019) and in outcrop (Backert et al., 2010;Gobo et al., 2014;Postma, 1984;Rubi et al., 2018). Similarly, they could also represent longer-lived tributaries into the larger N-S-directed erosional features. The current dataset is not sufficient to conclude whether the N-S-trending erosional feature (Area E; Figures 4 and 5b) formed concomitant with the clinoform package or if it is a younger and distinct, separate feature. However, the main erosional surface of the BCU is locally underlain by slightly older, intra-Late Jurassic erosive features within the northern most part of Depocentre E (Figures 5b and 7d) suggesting that the head of this feature may have initiated during the Late Jurassic, with minor periods of filling, prior increased erosion as part of a larger, subaqueous drainage system during the Cretaceous.

OF THE LATE JURASSIC RIFT TERRAIN
The Upper Jurassic clinoform packages on the Frøya High, particularly their topsets that were deposited in a near-horizontal coastal plain or shallow marine platform setting, may provide constraints on the amount of tilting that has occurred since Late Jurassic times. Figure 11 shows the depth-converted section rotated by 20° on the basis the western flanks of the clinoform packages are interpreted to be sub-horizontal topsets during the Late Jurassic. The resultant section (Figure 11c) represents an approximation of the attitude of sedimentary rocks of the Viking Group on the Frøya High at the time of deposition (Figure 12). Our simple approach does not incorporate flexure and compaction and is less accurate away from the anchor point, Viking Group wedge 'B' in Figure 5a, and especially where uplift would be greater, closer to the KFC. Additionally, the Top Viking Group/ BCU structure map is a composite surface that in places is conformable (e.g. near 6306/6-1 and 6306/9-1), but in others reflects substantial erosion (e.g. areas marked '1', '2' and '3' in Figure 12). In areas where the Viking Group was partially removed during the post-rift phase the surface contains topographic elements that post-date the Jurassic and so may not reflect the Late Jurassic configuration of the surface within those specific areas. Despite these shortcomings of our simple approach, the restoration of Figures 11  and 12 performs well in the eastern flank of the Frøya High and within the clinoform packages where there is limited post-Late Jurassic erosion. The restoration suggests that a prominent irregularity on the basement contact immediately west of the Almond Basin ('y' in Figure 11b) formed a topographic high during late Jurassic times (Figure 11c). The thickness maps of Figure 5 reveal that this topographic high corresponds to an area of nondeposition during the Late Jurassic ( Figure 5a) and to some extent also during the Early Cretaceous (Figure 5b). Together, these observations suggest that this north-south trending basement ridge likely represented the apex of the Frøya High during the Late Jurassic seen in both the Viking Group thickness map and the restored Top Viking Group structure map (Figures 5a and 12). We interpret this ridge to represent the drainage divide of the Frøya High as it existed during Late Jurassic and possibly Early Cretaceous times (Figure 3).

PALAEODRAINAGE AND PALAEOGEOGRAPHY OF THE FRØYA HIGH
The interpreted prograding shoreline-delta system developed east of, and parallel to, the eastern margin of the Almond Basin. The structural restoration of Figure 12 suggests that the Almond basin was positioned topographically higher than the clinoform packages during Late Jurassic times and was likely subaerially exposed when the clinoform packages were prograding. In addition, the interpreted palaeodrainage divide of the Late Jurassic was located parallel to and immediately west of the Almond Basin. These interpretations together allow constraint on the position and geometry of the palaeocatchments, which fed the clinoform packages, and broadly encompass the area 'O' in the Upper Jurassic thickness map of Figure 5a.
Each of the distinct Upper Jurassic clinoform packages (A, B, C in Figures 5a and 13) is connected to the  (Figure 5a). Each of these narrow corridors is separated by areas where the Upper Jurassic stratigraphy is extremely thin (<20 m) or largely absent (I, II and III in Figure 5a). In these areas F I G U R E 1 1 Cross sections demonstrating a simple restoration of the Top Viking Group/BCU structure map. The surface of Figure 4c is depth converted and rotated, so that the areas interpreted to represent horizontal topsets of the deltaic clinoform packages (a-d) (Figure 12). Location in Figure 1a. (I, II and III in Figure 5a) the strong negative acoustic impedance, linked to organic-rich shales typical of the Top Viking Group is still preserved (e.g. intersection with Area II in Figure 7b) and so indicates these areas do not simply represent enhanced Early Cretaceous erosion unlike deeply eroded areas of Viking Group stratigraphy. We interpret the thickness patterns of the Viking Group to reflect that the geometry of palaeocatchment area 'O' (Figure 5a) was likely subdivided into a series of parallel drainage catchments with SE-flowing river systems that debouched to the southeast and supplied sediment to individual point-sourced fan deltas between areas I, II and III (Figures 5a and 13). The area occupied by the palaeocatchment on the restored structure map of Figure 12 is affected by Early Cretaceous erosion, which renders it inadequate for mapping more subtle features of Late Jurassic such as individual drainage profiles. However, the restoration does provide some constraints on overall catchment geometries.
Measuring from the maximum eastern (basinward) extent of the Viking Group thickness anomalies to the interpreted drainage divide provides an approximate maximum catchment length of 12.5, 16.9 and 18.3 km for clinoform packages A, B and C, respectively, with a mean catchment length of 15.9 km (Figures S1 and S2). The catchment for clinoform package D is only partly covered by the dataset and so the length is difficult to constrain. The total basinward extent of the Viking Group thickness anomaly probably overestimates the catchment length as it includes a portion of subaerial topset, subaqueous foreset and bottomset, which is not part of the subaerial catchment (Hovius, 1996). Alternatively, the location of catchment outlets can be estimated from the isopach map as the region where there is an abrupt widening and onset of a fan-like geometry of the thickness anomalies. Using the estimation of the catchment outlet provides a minimum estimate of the catchment length, which are 8.0, 11.3 and 12.1 km for clinoform packages A, B and C, respectively, giving an average length of 10.5 km and indicating an average clinoform package length of 5.2 km ( Figures S1 and S2). These dimensions are in keeping with ancient and modern deltaic clinoforms observed in the Corinth Rift (e.g. Barrett et al., 2019) and measurements in compiled database studies for sand-prone deltaic clinoforms (Patruno et al., 2015).
The spacing between catchments can be roughly estimated using the small, 1-5 km long, ~100 m high, relative highs (I, II and III in Figures 5a and 13a) and assuming that the catchment outlets were located centrally between them. This analysis gives catchment spacings of 5.7 km between C and B, and 7.6 km between B and A, averaging 6.7 or 6.3 km if the Viking Group thickness anomaly (Wedge D) north of Wedge A, to the west of 6306/6-1 is also included (Figures S1 and S2). The spacing ratio of catchment length to spacing averages 1.69 using minimum length estimates and 2.4 using maximum length estimates, in keeping with catchment area morphometrics for active basins in other studies (2.07- Hovius, 1996;2.5-Talling et al., 1997;2.48 Sømme et al., 2010). The area of the catchments ranges between 44-104 km 2 , which is typical for small catchments common in active rift margins (e.g. Eliet & Gawthorpe, 1995). Characterisation of the Frøya High bedrock stratigraphy highlights that Late Jurassic catchments consisted of relatively easily erodible sedimentary rocks of Upper Palaeozoic-Lower Mesozoic age within the Almond Basin, as opposed to the crystalline rocks elsewhere, outside the Almond Basin, on the southern Frøya High (Jones et al., 2021;Muñoz-Barrera et al., 2020. We conclude on the basis of spatial patterns of Viking Group thickness, the position of a NE-SW trending ridge following restoration and comparison with other rift system catchment morphometrics, that the interpreted Upper Jurassic clastic shoreline system was supplied by at least four distinct drainage basins, each around ~10 km, long spaced 5.5-7.5 km along the dip slope of the Frøya High. Each drainage basin was preferentially located on exposed pre-Middle Jurassic sedimentary rocks of the Almond Basin, which provided easily erodible material feed clinoform packages that prograded eastward into the Froan Basin.

EVOLUTION OF THE FRØYA HIGH
The erosional features of the Top Viking Group/BCU surface removed substantial portions of the Upper Jurassic stratigraphy (e.g. Area 1-5 in Figures 12 and 13b). The timing of this erosion is poorly constrained and it is possible that it could have commenced during the end of the latest Jurassic (e.g. sub-BCU erosive features in Figure 7d). Areas 1-3 exhibit a broadly N-S-orientation, compared with E-W orientation of the Late Jurassic drainage catchments (Figures 12 and 13). Moreover, these areas seem to represent catchments that drain towards the southwestern edge of the Frøya High. Areas 4 and 5, however, have the same location and orientation as their corresponding Late Jurassic erosional features (Figures 12 and 13). It thus follows that those catchments that drained westward into the Møre Basin were not rearranged during the transition from the Late Jurassic to the Early Cretaceous. Those Jurassic catchments (Areas 1-3) that used to be contained within the east-facing dip slope, however, underwent significant change. A clear indication of the mechanism behind this rearrangement of drainage on the dip slope is suggested by Area 3 (Figure 13b), where headward erosion from one of the fault-scarp-bounded catchments (3 in Figures 12 and 13b) appears to have captured a large portion of the catchment that was previously draining eastward. The final form of the erosive features (e.g. 1-5 in Figure 13b) is interpreted to be of Early Cretaceous age as large portions of Upper Jurassic stratigraphy that covered the Almond Basin, likely deposited during transgression of the Jurassic drainage basin ('O' in Figure 5), were eroded by the younger, southward draining catchment. We propose that this must have occurred as the footwall scarp catchments along the KFC continued to grow by headward erosion.
Although the relatively deep, southward-directed canyon immediately east of the clinoform package (Area 2, in Figure 12) extends south beyond the study area it seems plausible to be the result of a similar drainage capture event that could have occurred farther to the southeast along the KFC fault scarp. Observations of degradation of the delta foresets and sub-BCU erosion in the northern part of this erosive feature (e.g. Figure 7d) indicate that the Area 2 canyon may have begun forming whilst the clinoform packages were active during the Latest Jurassic. Here, degradation of the clinoform delta front may have facilitated sediment transport into and along the basin floor in an axial channel system not dissimilar to axial channels observed in front of modern fan deltas (e.g. Beckers et al., 2018;Gales et al., 2019;Gawthorpe et al., 2018;McNeill et al., 2005;Prior & Bornhold, 1989). However, the data are not able to resolve how much of this erosion substantially post-dates the emplacement of the clinoform packages (i.e. during the Early Cretaceous) and are formed by more widespread, headward, subaqueous erosion of the then submerged Frøya High flank. Nevertheless, the isolated positions of Cromer Knoll Group thicknesses and onlap of the Cromer Knoll Group onto the BCU suggest that the topography of Area 2 within the canyon was in place by the end of the Early Cretaceous (Figures 4, 7 and 13).
The canyon at Area 2 likely formed a substantial topographic/bathymetric low within which the Cromer Knoll Group was deposited ( Figure 13) and may well have formed through a composite history of both Late Jurassic and Early Cretaceous erosion. The Lower Cretaceous Cromer Knoll Group in the study area consists of continuous, subparallel strata that onlaps the BCU with some minor concave-up geometries within the broad N-S orientated erosional feature (Area E, Figures 4, 6 and 7d). The lack of any major constructional features along the flank of the Frøya High (cf. Upper Jurassic clinoform packages) suggests that the Frøya High depositional systems, which routed sediment towards the Froan Basin likely shutdown and that sediment routing, were largely focussed to the south and west of the Frøya High through the Early Cretaceous. A termination of sediment supply from region O ( Figure 5) is in agreement with the gradual onlap of the Frøya High by the Cromer Knoll Group, recording a relatively fast transgression of the Frøya high at the end of the Jurassic and into the Early Cretaceous (Figures 3c and 9; Brekke, 2000;Bell et al., 2014;Jones et al., 2021). 9 | DISCUSSION 9.1 | Alternative interpretations of the studied clinoform packages: The Rogn formation as a connected, constructional coastline on the Frøya high Our interpretation of the Upper Jurassic packages on the south-eastern part of the Frøya High as deltaic clinoform packages is different from earlier interpretations. Chiarella et al. (2020) propose the same sand-prone packages within the Rogn Formation to be interpreted as coastal sand ridges sourced and constructed solely through longshore currents, using comparisons between the core from the well 6306/6-1 and the Draugen field 100-125 km to the northeast (also described by Van der Zwaan, 1990). Deltaic clinoforms are our preferred interpretation on account of the lobate, convex to the east plan view geometry, eastward progradation, consistent with an overall upward-coarsening, poorly sorted, coarse-grained foreset-dominated package in 6306/9-1 (Figure 9), typical of fan deltas (e.g. Nemec, 1990;Rohais et al., 2008). Furthermore, the location of the clinoform packages down-dip of a prominent, erosional, drainage catchmentliked geometries, floored by highly erodible sedimentary deposits of the Almond Basin implies the bulk of sediment was deposited within deltas at the coastline with only minor reworking and northward transport of sediment within delta topset deposits (e.g. Figures 9, 13, and 14). The clinoform packages all exhibit a consistent progradational architecture from west to east, with parallel topsets overlying or connected to generally east-dipping sigmoidal clinoforms when Upper Mesozoic and Cenozoic rotation is removed by flattening onto a Top Shetland Group horizon ( Figure 7). Chiarella et al. (2020) do refer to westward dipping foresets in the northernmost post of the dataset, immediately west of 6306/6-1. However, we observe only a single westward dipping surface (Figure 7d), which ties down-dip into the large N-S trending erosional feature (e.g. Depocentre E, Figure 5). We therefore interpret this westward dipping reflector to be the margin of the up-dip head of this erosional feature that post-dates the delta itself. The consistent eastward progradation is more compatible with that of a radial fan delta than with an offshore sand ridge, which would likely show coastline parallel progradation. Whilst progradational clinoforms do form within offshore sand ridges (e.g. Berne et al., 1998), their angle is typically <10°, and foreset heights rarely exceed 30 m (Chiarella et al., 2020), an order of magnitude smaller than the maximum height of those observed in the Frøya High study area. The abundance of bioturbated, macro-fossiliferous, plant-material-rich facies of the Rogn Formation described in 6306/6-1 by Chiarella et al. (2020) and used to ascribe a sand-ridge interpretation are not necessarily unique to that depositional environment. Bioturbation, macro-fossil and plant-material-rich sandstones are common in a broad range of shoreline and near-shoreline depositional environments and are especially common in wave-or tide-modified delta fronts (Anell et al., 2021;Gastaldo & Huc, 1992;Rossi & Steel, 2016).
Our interpretation of the studied Upper Jurassic clinoform packages as a series of fan deltas has implications for the overall depositional setting and sediment flux across the eastern flank of the Frøya High during the Late Jurassic. Figures 13 and 14 illustrate a new, depositional model that shows a coastline along the eastern side of the Frøya High that is more constructive and dominated by transverse sediment input, rather than axial reworking that has dominated previous models driven by observations from the Draugen sand ridge (e.g. Chiarella et al., 2020;Van der Zwaan, 1990). Palaeocurrent measurements from 6306/9-1 do suggest a component of longshore reworking near the top of the Rogn Formation; however, we interpret F I G U R E 1 4 A new, simple palaeogeographical model for the Late Jurassic deposition along the south-eastern margin of the Frøya High. The shoreline is supplied with abundant clastic material in the south, and potentially in the middle, and a longshore current connects it to the Draugen ridge that may represent a spit system rather than a detached sand ridge. this to be limited to reworking within the topsets of deltas, with the large thicknesses of sediment within the proposed deltas suggesting considerable volumes of sediment are stored within the deltas rather than transported further northwards.
Whilst the existence of shallow marine systems flanking the transition between the Frøya High and Froan Basin has long been speculated on the basis of observations from the Draugen Field (Chiarella et al., 2020;Van der Zwaan, 1990), we speculate further that the Draugen sand ridge may be more intimately related to the regressive, deltaic shoreline documented here. Existing interpretations for sedimentation of the Draugen Ridge lacked a demonstrable sedimentary input for sediment reworked axially from the southwest. The proto-Norwegian Sea existed as a narrow seaway located above 30 degrees latitude from the early through to the Late Jurassic. Palaeoclimate interpretations for the Late Jurassic and Early Cretaceous of the North Sea and NW Europe (e.g. Abbink et al., 2001;Mutterlose et al., 2003) highlight that the prevailing current direction was controlled by the northern hemisphere westerlies, i.e. directed to the northeast. These current directions are consistent with the dune migration direction proposed by Chiarella et al. (2020) for the asymmetry of the Kimmeridgian Rogn Formation at the Draugen field, and with Early Jurassic Ilje Formation shallow marine bar progradation, and longshore sediment redistribution in the Halten Terrace (Martinius et al., 2001). As a result of these westerlies, the northwestern margin of the Frøya High likely received a strong northeastern current as the southern part of the Sklinna Ridge was submerged and formed an opening to the proto-Norwegian Sea to the west (Bell et al., 2014;Elliott et al., 2017). Similar currents along the eastern flank of the Frøya High were also likely in operation due to the same prevailing current direction, which may have been enhanced through the interaction of the several exposed islands in the narrow seaway of the proto-Norwegian sea ( Figure 14). Petrographic information of the Kimmeridgian Rogn Formation at the Draugen ridge (Chiarella et al., 2020;Van der Zwaan, 1990) and broader Møre-Trøndelag area (Mørk & Johnsen, 2005) indicate a provenance with minor reworking from local granitic and Mesozoic-sediment covered palaeohighs such as the Frøya High to the south and west of the Draugen Ridge and Froan Basin (Jongepier et al., 1996;Mørk & Johnsen, 2005). This petrographic signature is consistent with a redistribution of sediment along strike in the shoreline region of the Frøya High-Froan basin margin, with the confluence of these currents meeting at the Draugen Ridge, on the eastern tip of the Vingleia Fault ( Figure 14). The deltaic clinoform systems recognised in this study in the southern part of the Frøya High likely provide the primary sedimentary input for the more minor amount of sediment, which may bereworked along the Frøya High-Froan basin margin. The description of the Draugen ridge presented by Chiarella et al. (2020) and Van der Zwaan (1990) along with our model of redistribution is consistent with the prediction by Nielsen and Johannessen (2009) of how spit systems will often locate at the confluence of meteorological oceanic currents around a landmass, along-strike from major sediment input points, where a majority of sediment remains. The stratigraphic architecture recognised by Chiarella et al. (2020), with a large subaqueous, mud-dominated platform on which upwardcoarsening units from subaerial, high-energy environments are deposited also bear similarities with modern and recent spit environments documented in Nielsen and Johannessen (2009). 9.2 | Implications for models of clastic shoreline sedimentation on rift-related dip slopes 9.2.1 | Impact of hinterland characteristics and source-to-sink configuration on depositional systems around intrabasinal highs In the case of the Frøya High, the nature of the bounding structures and resultant physiography of the Late Jurassic landscape can produce substantial along-strike variability in the nature of the coastline. This case study demonstrates that if a relatively sizeable source for sediment is exhumed during the evolution of a rift, even in an intra-basinal location, constructive coastal environments may develop locally on dip slopes, especially downslope of easily erodible sedimentary bedrock. At the southern edge of the Frøya High, a relatively broad, back-tilted, footwall dip slope consisted of an easily erodible exhumed sedimentary basin (the Almond Basin) that supplied sediment to a series of closely spaced fan deltas and resulted in a regressive coastline. Less than 100 km to the north, where the Frøya high is significantly narrower and consists of resistant, crystalline bedrock, sediment supply is likely dependent on longshore drift (Chiarella et al., 2020;Van der Zwaan, 1990). The longshore drift-dominated coastlines in this region may be dominantly transgressive compared with more progradational or aggradational time-equivalent systems along-strike to the southwest. Muravchik et al. (2018) highlight the co-existence of sediment-starved dip slope and delta-fed dip slopes over strike distances of 10-20 km from the El-Qaa fault block of the Suez Rift. Muravchik et al. (2020) suggest that the steepest structural gradients and greatest uplift, towards the centre of fault segments, drive the location of primary sediment input points on dip slopes, whereas shallow structural gradients around fault tips are more likely to be easily transgressed by rising sea level and sediment starved. Our findings indicate that, in addition to structural changes, the width and bedrock composition of the dip slope are important factors to consider in controlling sediment supply, the nature of the dip slope coastal plain and shallow marine regime. Narrow exposed zones of dip slopes, without the development of drainage catchments, may not be able to contribute significant sediment yield in order to build constructive coastlines (Figure 15a). Broader dip slopes can develop larger catchments and potentially connect to greater altitudinal differences from headwaters to drainage outlet, with consequently greater discharge and erosional capability from F I G U R E 1 5 Conceptual cartoons for evolutionary models of dip slope shoreline systems in (a) A footwall uplift-driven dip slope shoreline system. Here, progressive uplift in the footwall of a major fault drives sediment flux (T1-T2) but also induces headward erosion from other catchments, which capture dip slope drainage (T3). (b) A hangingwall subsidence-driven dip slope shoreline systems. Here, progressive subsidence in the hangingwall of a major fault drives a structural gradient to produce drainage catchments, which are gradually transgressed (T2) and recorded as protracted shutdowns of shoreline systems (T3). steeper drainage profiles (Nyberg et al., 2021;Romans et al., 2016;Sømme et al., 2009). dip slope Nevertheless, the sediment yield is likely to be relatively small given the shallow angle of dip slopes (0-7°, Muravchik et al., 2018) compared with steeper catchment systems developed on the footwall scarp of normal. Slower subsidence rates, shallow gradients and shallow water depths make dip slope systems prone to be constructional depositional systems unlike steep, deep fault-attached systems in the immediate hangingwall of major normal faults (Gawthorpe et al., 1994;Rapozo et al., 2021;Ravnås & Steel, 1998;Smyrak-Sikora et al., 2021). For example, the Viking Group in the hangingwall of the major faults (KFC and VFC) that bound the Frøya High are dominated by deepwater fans and bypassed fault scarps with limited capacity for self-development or shoreline sediment storage (Elliott et al., 2017;Jones et al., 2021). The tectonostratigraphic setting of constructive shorelines on the low gradient, back-tilted footwall dip slopes may record substantially different stratigraphic architecture and depositional stacking patterns due to their increased sensitivity to eustatic base-level changes compared with depositional systems in the immediate hangingwall of a major normal fault (Barrett et al., 2018;Gawthorpe et al., 1994;Gawthorpe et al., 2017;Henstra et al., 2017;Jackson et al., 2005;Leeder & Jackson, 1993;Ravnås & Steel, 1998;Smyrak-Sikora et al., 2021) Jackson et al., 2005;Muravchik et al., 2018); Snorre-H area and, Central Graben of the North Sea (Nøttvedt et al., 2000); Upper Heather Member, Oseberg Fault Block (Ravnås et al., 1997;Ravnås & Steel, 1998), all exhibit a net-transgressive character from drowning of the dip slope similar to the capping of the deltas here by a transgressive section (Figure 9). Large-scale, terminal flooding occurs because of continued rotation and net subsidence of the dip slope in response to either to background subsidence of the margin and cessation of fault or enhanced activity in the hangingwall of a nearby fault (e.g. Ravnås et al., 1997;Ravnås & Steel, 1998). Unlike other published examples, all the clinoforms observed on the Frøya High show strongly progradational character, and we observe no backstepping of deltaic units at the seismic scale prior to their termination. Observations in 6306/9-1 demonstrate a well-developed delta topset, which is abruptly flooded with a thin transgressive-lag overlain by offshore mudstones of the Spekk Formation, rather than more gradual backstepping (Figure 9). This lack of protracted, terminal retrogradation suggests flooding occurred abruptly due to rapid sea-level rise or drastic sediment supply reduction, which would be counterintuitive against the ongoing uplift-induced steepening of the catchment headwaters. The Abrupt transgression of the Frøya dip slope systems may be at least partly explained through the drainage capture from headward erosion by catchments on the Klakk Fault Complex footwall crest (Figure 15a). The Early Cretaceous timing of drainage capture to south/southwest draining catchments is simultaneous with previously documented acceleration and localisation of deformation on the Klakk Fault Complex due to its linkage to a major detachment in the lower crust (Bell et al., 2014;Muñoz-Barrera et al., 2020;Peron-Pinvidic et al., 2013). The process of drainage capture by footwall scarp drainage and reduction in dip slope catchment area may be a common rearrangement where displacement on the controlling normal fault continues or accelerates rather than transferring to other structures (Leeder et al., 2005;Privat et al., 2021).
To maintain progradational architectures without retrogradation during net subsidence, however, requires substantial sediment flux. The balance of sediment supply to accommodation for dip slope depositional systems will be influenced by the rotation of the fault block, uplift of the catchment to steepen drainage profiles, the gradient of the subsiding (depositional) part of the dip slope and any changes in sea-level (Gawthorpe et al., 1994;Ravnås et al., 1997;Ravnås & Steel, 1998). Dip slopes in rift basins can occur through two relationships to an active normal fault; rotation in the footwall of a major active fault (Figure 15a) or rotation in the hangingwall of a major active fault (Figure 15b). In a footwall uplift-driven system such as the Frøya High, we observe strongly progradational architectures on the dip slope prior to an abrupt termination of the system from drainage capture and flooding rather than the preservation of a gradual retrogradation and transgression of the clinoform packages. The existence of deltaic systems on the Frøya High in a depocenter approximately 20 km from the drainage divide, and 30 km from the Klakk Fault Complex, is consistent with a change to net subsidence, due to contributions from background subsidence, at half the wavelength of flexure of most large normal faults (40-60 km; Morley & Lambiase, 1995;Armijo et al., 1996;Fernández-Blanco et al., 2020). Given the decay of uplift to subsidence away from a bounding fault across a dip slope, the uplift of a dip sope catchment closer to the fault is likely to be greater than the subsidence experienced in the dip slope further from the fault. The resultant uplift-induced steepening of the drainage profile in the upper reaches of catchments is therefore likely to allow an increase in sediment flux, which can exceed the accommodation generated by subsidence to produce the strongly aggradational to progradational character observed on the Frøya High dip slope (Figure 15a). Conversely, in a hangingwall subsidence-driven setting (e.g. Alkyonides Gulf-Leeder et al., 2005;El-Qaa Fault Block-Muravchik et al., 2018) (Figure 15b), subsidence is likely to be greater in the depocentre than the uplift in the catchment feeding the dip slope depositional systems and uplift and steepening of the catchments may be very minor, or the catchment may even undergo net subsidence (Pechlivanidou et al., 2019). As a result, even coastal systems with a fluvial input become accommodation-dominated and are easily transgressed by subsidence-induced sea-level increases producing the retrograding stacking patterns observed in hangingwall subsidence-driven dip slope systems such as those seen in the dip slope systems of the Alkyonides Gulf (Leeder et al., 2005), Oseberg Fault Block (Ravnås et al., 1997;Ravnås & Steel, 1998) and El-Qaa and Hammam Faraun Fault Blocks (Jackson et al., 2005;Muravchik et al., 2018). Temporal changes in fault activity (similar to the high and low 'tilt rates' of Ravnås & Steel, 1998) and interactions with nontectonic changes in sea-level or sediment flux have the capacity to alter this motif; however, the differences between hangingwall and footwall driven dip slope systems highlights that the location of synchronous fault activity is a key controlling parameter on evolution and resultant stacking patterns.

| CONCLUSIONS
Detailed seismic mapping and analysis of borehole data within the Viking Group in the footwall dip slope of the Klakk Fault Complex has revealed a constructive, deltadominated shoreline system on the eastern flank of the Frøya High. These sedimentary systems are sourced from eroded material from the eastern side of a drainage divide in the footwall of the Klakk Fault Complex that was transported eastwards down the back-tilted, footwall dip slope into the Froan Basin. This case study provides an example of the facies and stratigraphic architecture of dip slope depositional systems within rift settings and highlights spatial variations in dip slope sedimentation over 100 km along strike.
The example on the Frøya High demonstrates the potential for intra-basinal highs to provide substantial sediment sources not only into the hangingwall of major faults but also across the back-tilted dip slope in their footwall. Comparison with previous interpretations, and other examples of syn-rift dip slope systems, highlight the likely transient nature of coarse-grained systems on dip slopes as they respond to steepening of the headwaters of feeding catchments, which can drive high sediment flux and progradation despite ongoing subsidence further down the dip slope. We hypothesise that footwall uplift-driven systems are likely to be more prone to progradation followed by abrupt shutdown resulting from drainage capture by footwall scarp drainage networks. By contrast, hangingwall-driven systems are more prone to gradual transgression and form overall retrograding systems. Integration with regional palaeogeography highlights that the deltaic packages may be important for determining the location, sediment budget and calibre of sediment supplied to longshore shallow marine systems such as spits and sand ridges such as the Draugen Ridge. The study ultimately highlights the major progradation of dip slope systems on the back-tilted footwall dip slopes of intra-basinal highs bound by high displacement normal faults, especially where bedrock is composed of easily erodible material. This is in contrast to shallow gradient low-displacement dip slope systems or those generated in the hangingwall of large structures, which generally have more limited progradation, and preserve strong retrogradational stacking patterns.