Abstract

A paucity of studies on mud-rich basin slope successions has resulted in a significant gap in our sedimentological understanding in these settings. Here, macro- and micro-scale analysis of mudstone composition, texture and organic matter was undertaken on a continuous core through a mud-dominated slope succession from the Marl Hill area in the Carboniferous Bowland Basin. Six lithofacies, all dominated by turbidites and debrites, combine into three basin slope facies associations: sediment-starved slope, slope dominated by low-density turbidites and slope dominated by debrites. Variation in slope sedimentation was a function of relative sea-level change, with the sediment-starved slope occurring during maximum flooding of the contemporaneous shelf, and the transition towards a slope dominated by turbidites and then debrites occurring during normal or forced shoreline progradation towards the shelf margin. The sediment-starved slope succession is dominated by Type II kerogen, whereas the slope dominated by low-density turbidites is dominated by Type III kerogen. This study suggests that mud-dominated lower slope settings are largely active depositional sites, with consistent evidence for sediment traction. Additionally, the composition and texture of basin slope mudstones, as well as organic content, vary predictably as a function of shelf processes linked to relative sea-level change.

The recent development of unconventional hydrocarbon systems that target source rock reservoirs (organic-rich mudstones acting as both source and reservoir) has led to a re-evaluation of fundamental geological processes operating in fine-grained sedimentary systems (e.g. Ross & Bustin 2009; Aplin & Macquaker 2011; Jarvie 2012; Hart et al. 2013; Taylor & Macquaker 2014). Depositional and post-depositional processes within fine-grained sedimentary systems are complex and control the distribution of rock properties, both spatially and temporally, which are key considerations for the appraisal, exploration and production of source rock reservoirs (e.g. Jarvie et al. 2007; Loucks et al. 2009, 2012; Charpentier & Cook 2011; Andrews 2013). The style of deposition (e.g. the amount of organic matter flux, the type of organic matter present, the detrital grain assemblage, the sediment accumulation rate) is therefore critical in determining the characteristics and variability within any fine-grained sedimentary succession (e.g. Macquaker et al. 2007; Milliken et al. 2012; Hart et al. 2013; Milliken 2014; Lazar et al. 2015a; Schieber 2016a). Previous work on mud depositional settings has principally focused on processes acting on shelf (e.g. Schieber 2007; Macquaker et al. 2010; Schieber et al. 2010; Davies et al. 2012; Bohacs et al. 2014) or basin floor successions (e.g. Könitzer et al. 2014; Schieber 2016b). By focusing on the sedimentology and the resulting mudstone textures and variability, this paper aims to address the knowledge gap of processes acting on mud-dominated slope successions. The specific questions addressed in this study are as follows.

  1. What are the dominant modes of deposition active within a mud-rich slope succession?

  2. Is slope sedimentation inherently linked to contemporaneous shelf sedimentation?

  3. Is there a discernible relationship between the type and content of the organic matter of facies and facies associations within a slope succession?

These detailed sedimentological observations have been made on a continuous core (Marl Hill: MHD11; British Geological Survey reference number SD64NE/20), that preserves the transition from carbonate-dominated to clastic-dominated mudstone deposition, within a marginal to basinal mudstone succession, from an active exploration play, the Bowland Shale Formation (Carboniferous, UK).

Geological setting of the Bowland Basin

During the Viséan and Namurian stages of the Carboniferous (347 – 318 Ma; Figs 1 and 2), a series of interconnected half-graben basins within the Pennine Province (northern England, UK) formed part of a shallow epicontinental seaway that stretched across the Laurussian continent, with long-lived palaeogeographical barriers to the north (Southern Uplands High) and to the south (Wales–London–Brabant High; Woodcock & Strachan 2002; Davies 2008; Waters et al. 2009).

The interconnected half-grabens of the Pennine Province formed as a result of back-arc rift extension related to the Variscan Orogeny (410 – 280 Ma), with a strong NW–SE and NE–SW orientation inherited from pre-existing Caledonian structures (Fraser & Gawthorpe 1990). The Bowland Basin, the focus of this study, is a complex NE–SW-trending fault-controlled half-graben within the larger Craven Basin (Fig. 1). The margins of the Bowland Basin are defined by the Bowland Line to the north, the Craven Fault System to the NE (forming the boundary with the Askrigg Block ‘high’), and the Pendle Fault to the south (forming the boundary with the Central Lancashire High; Gawthorpe 1987; Arthurton et al. 1988; Riley 1990; Fraser & Gawthorpe 2003). Initiation of the Bowland Basin occurred in the latest Devonian to Tournaisian, with two further distinct episodes of rifting occurring during the Chadian–Arundian and Asbian–Brigantian (Gawthorpe 1987). Within the Bowland Basin a number of lesser depositional highs are present, which were primarily controlled by SW–NE-trending reactivated Caledonian basement faults (Gawthorpe 1987; Riley 1990; Waters et al. 2009). Depositional environments in the Bowland Basin evolved from a carbonate ramp to carbonate-rimmed shelf, with a distinct shelf–slope–basin-floor morphology through the Tournaisian to Visean (Gawthorpe 1987). Clastic deep-water sedimentation began to replace carbonate sedimentation from the late Visean onwards (Aitkenhead et al. 1992; Brandon et al. 1998; Kirby et al. 2000; Kane 2010), with progradation of the Millstone Grit deltas into the study area from the north during the late Pendleian. The Bowland Shale Formation typically has a thickness of 120 – 620 m (Waters et al. 2009), thickening southwestward towards the Pendle Fault (Evans & Kirby 1999). Within the Bowland Basin the deposition of thick mud-rich successions has been interpreted as the result of hemipelagic settling from suspension, largely below the storm wave-base, interrupted by periodic input from carbonate-rich and siliciclastic-rich turbidites and possibly storm events (e.g. Aitkenhead et al. 1992; Waters et al. 2009).

Stratigraphy of Marl Hill borehole MHD11

The Marl Hill borehole MHD11 core [SD 6833 4683] was collected by BP Mineral International Ltd for the purpose of prospecting for Pb–Zn mineralization associated with the proximal Clitheroe Fault Zone. The core is currently stored by the British Geological Survey at Keyworth (UK). The core penetrates a 300.58 m succession consisting of the following stratigraphic intervals: Worston Shale Group (consisting of the Hodder Mudstone Formation, Limekiln Wood Limestone Formation and Hodderense Limestone Formation) and Bowland Shale Group (consisting of the Pendleside Limestone Formation, Lower Bowland Shale Formation and Upper Bowland Shale Formation; Fig. 2; Aitkenhead et al. 1992). Based on a revised stratigraphic framework for the Carboniferous rocks of Britain (Waters et al. 2009, 2011), this interval equates to the Craven Group, consisting of the Hodder Mudstone Formation, Hodderense Limestone Formation, Pendleside Limestone Formation and Bowland Shale Formation. It is useful to refer to the former stratigraphic nomenclature as this allows correlation with older, established stratigraphies such as the seismic stratigraphy of Fraser & Gawthorpe (2003).Three marine bands are present within the cored Bowland Shale interval, the B2bGoniatites globostriatus (Beyrichoceras biozone, Upper Asbian), P2Sudeticeras (Upper Brigantian) and E1aEmstites (Cravenoceras) leion (Lower Pendleian) (Figs 2 and 3; Riley 1988; Aitkenhead et al. 1992). The focus of this study is the succession from 5 to 125 m consisting of the Hodder Mudstone Formation, Hodderense Limestone Formation, Pendleside Limestone Formation and Bowland Shale Formation. The core is slabbed, and recovery of the core in the interval from 5 to 125 m is c. 95%.

Methods

The illuminated, wetted core was logged for macroscopically visible textural (grain-size) and compositional changes, sedimentary features and fossil content at 10 cm resolution. The mudstone was classified according to the scheme of Lazar et al. (2015b). That is, sediments consisting of greater than two-thirds of grains <8 µm are defined as claystones, those with grains of 8 to < 32 µm are defined as mudstones and those containing two-thirds of grains of 32 to < 62.5 µm are defined as siltstones. The compositional modifier ‘argillaceous', ‘calcareous' or ‘siliceous' is added where >50% of the material comprises clay, carbonate or quartz, respectively. Compositional proportions for all scales (e.g. at core, scanned, optical microscope and SEM scale) were assessed using visual estimates. The core was subdivided into lithofacies according to these macroscale observations, and 36 samples were collected to prepare thin sections for identification of micro-scale features within these facies.

A total of 82 samples were taken across the cored interval, with 75 from the Bowland Shale Formation, using an adjusted uniform sampling method of one sample per metre of core with additional samples taken at points of interest. A further six spot samples were taken from the underlying Pendleside Limestone and Hodder Mudstone Formations (Figs 2 and 3). A total of 36 thin sections were chosen to represent all mudstone macrofacies identified at core scale, and prepared for polished (uncovered) thin sections normal to principal bedding orientation (Figs 3 and 4). All thin sections were scanned using an EPSON Perfection V600 Photo at 2400 d.p.i. resolution. Optical microscopy was undertaken using a Nikon Eclipse LV100NPOL microscope fitted with a Nikon DS-Fi2 camera. Scanning electron microscopy with an FEI XL30 Environmental scanning electron microscope with a backscattered electron detector attached was used for high-resolution imaging and compositional analysis.

Total organic carbon (TOC) was measured from 50 spot samples using a Leco CS230 carbon analyser, before analysis sample aliquots were de-calcified using 5M HCl and rinsed five times. Iron chip (0.7 g) and ‘Lecocell’ (tungsten) accelerants (c. 1.5 g) were then added to the crucibles. An induction furnace heated the samples to c. 1500°C in an oxygen atmosphere and the resulting CO2 was quantified using an IR detector. The instrument was calibrated using Leco supplied certified carbon standards appropriate to the carbon content of the samples being analysed; replicate and blank analyses were also carried out.

RockEval™ analysis was undertaken using a RockEval™ VI analyser. RockEval™ is a pyrolysis-based technique and results can be interpreted to indicate the type and maturity of organic matter and thus its petroleum generation potential (Espitalié et al. 1977). Between 50 and 100 mg of powdered whole-rock samples were pyrolysed in an inert helium atmosphere from 180 to 600°C, using a ramp rate of 25°C min−1. The mass of hydrocarbon vapours produced at 300°C (free hydrocarbons) was analysed using a flame ionizing detector (FID; S1 peak of the pyrogram). The temperature was then increased from 300 to 550°C to crack non-volatile organic matter (i.e. kerogen) (producing the S2 peak of the pyrogram, CO2 and water). The temperature at which the S2 peak reaches maximum generation is termed the Tmax and is a proxy for thermal maturity. The CO2 generated during thermal breakdown of kerogen is recorded as the S3 peak of the pyrogram. Further parameters were derived using the following equations (Tissot & Welte 1984): (1) hydrogen index, HI = (S2/TOC) × 100; (2) oxygen index, OI = (S3/TOC) × 100; (3) production index, PI = S1/(S1 + S2); (4) transformation index, TI = S1/TOC; (5) pyrolysate yield, PY = S1 + S2. All organic matter values either measured (using TOC or RockEval methods) or calculated reported in this study are present-day values.

Lithofacies analysis

The mud-rich succession in the Marl Hill core contains a total of six lithofacies comprising calcareous claystone, calcareous mudstone, argillaceous mudstone, calcareous siltstone, calcareous sandy mudstone and matrix-supported carbonate conglomerate (Facies 1–6 respectively; Figs 3 and 4). The mudstone lithofacies were defined according to their macroscopic texture and composition, following the classification scheme of Lazar et al. (2015b). Images of thin section scans, optical photomicrographs and SEM photomicrographs of each lithofacies are provided in Figure 4. The macroscopic and microscopic sedimentological, biogenic and diagenetic features of each lithofacies are summarized in Table 1. Interpreted depositional processes for each lithofacies are also given in Table 1.

Facies associations

Three facies associations have been identified, and are summarized below. They are here interpreted within the context of a shelf–slope–basin-floor system, previously determined for the study area during the mid-Carboniferous by Gawthorpe (1986, 1987).

Facies Association A: sediment-starved slope

This facies association comprises successions up to 58 m thick (Fig. 3), predominantly composed of calcareous mudstone (Facies 2) and argillaceous mudstone (Facies 3), intercalated with minor calcareous claystone (Facies 1), calcareous siltstone (Facies 4) and calcareous sandy mudstones (Facies 5). Facies 1, 4 and 5 occur as individual thin beds rarely thicker than 20 cm. This facies association represents deposition dominated by low-energy turbidites (predominantly T3–T8, with minor T0–T2 of Stow & Shanmugam 1980), in which the sediment was originally sourced from areas of carbonate productivity and clastic sediment supply. The presence of mudstone lenses in Facies 1, 2 and 3 suggests that mud-rich sediment was remobilized from a partially lithified substrate, transported via traction and incorporated within the succession. This facies association is the only one to contain calcareous claystone (Facies 1). Facies 1 represents the only facies with the potential for significant portions of the sediment to be derived from hemipelagic suspension, thus attesting to low rates of sediment supply within Facies Association A in contrast to other facies associations within the succession. The occurrence of thin beds of Facies 5 represents rare higher-energy debritic flows. The presence of soft-sediment folded beds in this facies association is indicative of sediment remobilization following deposition, caused by instability generated by gradients in a slope environment.

Facies Association B: slope dominated by low-density turbidites

This facies association comprises successions up to 15 m thick (Fig. 3), predominantly composed of calcareous mudstones (Facies 2), argillaceous mudstones (Facies 3), calcareous siltstones (Facies 4) and calcareous sandy mudstones (Facies 5). This facies association contrasts with Facies Association A by the absence of Facies 1, and by local presence of debrites (Facies 5). Deposition was predominantly via low-energy turbidites (T0–T8 of Stow & Shanmugam 1980). As with Facies Association A, the presence of mudstone lenses suggests that some of the sediment was partially lithified prior to deposition, and the presence of slumped horizons is indicative of deposition on a gradient. The absence of Facies 1 suggests that deposition rates were higher than in Facies Association A, resulting in the dilution of contributions from hemipelagic suspension settling. Overall, this facies association represents the lower portion of a slope that is actively receiving both carbonate and clastic sediment from the shelf.

Facies Association C: slope dominated by debrites

This facies association comprises successions up to 10 m thick (Fig. 3), predominantly composed of matrix-supported carbonate conglomerate (Facies 6) and calcareous mudstone (Facies 2). Facies Association C represents the alternation of deposition from moderate- to high-strength cohesive debris flows (DM−2, Talling et al. 2012) and low-energy turbidites (T3, T6 and T7; Stow & Shanmugam 1980). The polymict wackestone and floatstone lithology of the clasts in Facies 6 and the presence of biogenic material indicative of fully open-marine conditions suggest that the source of material was a low-energy carbonate-rich shelf below the storm wave base.

Facies association succession, depositional model and sequence stratigraphy

Metres 125 – 64 of the core are characterized by three whole coarsening-up successions. Metres 115 – 93 (sequence I; Fig. 3) record a change from a sediment-starved slope to a slope dominated by low-density turbidites. Metres 93 – 78 (sequence II; Fig. 3) record a change from a sediment-starved slope through a slope dominated by low-density turbidites to a slope dominated by debrites. Metres 78 – 64 (sequence III; Fig. 3) record a change from a sediment-starved slope to a slope dominated by low-density turbidites. The succession from 64 to 5 m (sequence IV; Fig. 3) is represented entirely by a sediment-starved slope.

The evidence of significant proportions of slumping and sliding as well as the occurrence of periodic inputs via low-density turbidites and debrites suggest that this is a slope succession (McIlreath & James 1978; Shanmugam 2000; Stow & Mayall 2000). Slope successions are inherently progradational (Patruno et al. 2015), with the thickness between basin floor sediment and shelf top sediments representing the (minimum) height of the slope clinoforms. No transition from basin floor to shelf top sediments is observed within the succession, suggesting that the slope clinoform was at least of the order of 120 m (accounting for decompaction). This thickness is within the range of 100 – 1000 m usually indicated for basin-margin clinoforms (Patruno et al. 2015), and, combined with the carbonate-rich nature of the sediment, strongly implies that it was a basin-margin, rather than delta-front, slope succession.

The lower three cycles (Cycles I–III) represent deposition on the slope with gradually increasing proximity to active sedimentation on the shelf. The initial slope conditions were relatively sediment-starved (Facies Association A; Fig. 5) because shorelines were distant from the shelf margin and active carbonate productivity and clastic sediment delivery occurred far away from the slope. Whilst on the slope, hemipelagic suspension settling of fine, organic-rich mud may have dominated. Episodic, storm-generated or river-effluent-generated density currents distributed fine-grained sediment across the shelf and onto the slope. Shelf-generated density currents may have bypassed the upper slope and deposited on the lower slope, or material deposited on the upper slope may have been remobilized via slumps and slides, transforming into turbidites and rare debrites, which deposited on the lower slope. We hypothesize that the lenses (shale lithic clasts) within the succession represent reworked, partially lithified upper slope sediments.

As shorelines prograded towards the shelf margin, the slope received increasing volumes of coarser sediment (Facies Association B; Fig. 5), largely delivered via storm-derived and river-effluent-derived density currents. In Cycle II, continued progradation of the shoreline resulted in oversteepening at the shelf margin and large-scale collapse of partially cemented carbonate-rich shelf sediments down the slope, resulting in the deposition of calcareous sandy mudstones and carbonate conglomerates on the lower slope (Facies Association C; Fig. 5). All three cycles are overlain sharply by sediment-starved slope sediments of the following cycle.

Cycles I–III, representing shoreline progradation followed by rapid regression across the contemporaneous shelf, occur through the Arundian to Asbian (Fig. 3; Waters et al. 2007), implying durations for each cycle of the order of c. 4 myr. Shoreline progradation occurs when sediment supply outpaces the rate at which accommodation is generated on the shelf, which may be during relative sea-level stillstand or rising relative sea-level (normal regression) or during relative sea-level fall (forced regression; Mitchum et al. 1977; Vail et al. 1977; Howell & Flint 1996; Catuneanu 2002; Gawthorpe et al. 2003). The duration of these cycles suggests that they represent sequence scale (third order; 1 – 2 myr) changes rather than stacked sequences (second order; 9 – 10 myr) or parasequence scale (fifth order; 0.01 – 0.02 myr; Mitchum & Van Wagoner 1991).

The upper succession (Cycle IV) from 64 to 5 m consists entirely of starved slope sediments (Facies Association A; Fig. 3), and represents an extended period of little sediment delivery to the slope, with carbonate and clastic sediments trapped on the shelf. The succession represents of the order of c. 5 myr of sedimentation, and implies the shelf was largely flooded for this period, with neither clastic nor carbonate shallow marine sedimentation being able to prograde across the shelf. The succession notably overlies the Goniatites globostriatus (B2b) ‘Marine Band’ (Fig. 3). Fossiliferous horizons, termed ‘marine bands', allow correlation across the Pennine Province (Ramsbottom 1973, 1978, 1979; Holdsworth & Collinson 1988; Waters & Condon 2012). Marine bands are interpreted to represent condensation associated with maximum flooding surfaces during major transgressions (third-order cycles; Mitchum & Van Wagoner 1991), and in the Namurian have been linked to possible fluctuations in Gondwanan ice-sheet dynamics and inferred Milankovitch orbital forcing (Isbell et al. 2003; Frank et al. 2008; Rygel et al. 2008; Stephenson et al. 2010; Waters & Condon 2012; Könitzer et al. 2014; Gross et al. 2015). The occurrence of the B2b at the base of Cycle IV can be used to infer that the cycles (I, II, III and IV) observed within the core could represent relative sea-level cycles similar to those observed by Ramsbottom (‘mesothems’ D2 to D6b; Ramsbottom 1973, 1981) and regressions observed by Ross & Ross (1987).

The variation in sedimentation style that is observed within the core at the boundary between Cycle III and Cycle IV (Fig. 3) is likely to have been effected by the changes in tectonism that the basin experienced at the time as well as the fluctuations in sea level (Fraser & Gawthorpe 2003; Haq & Schutter 2008). During EC3 (Chadian to Holkerian; Cycles I–III) the basin experienced a phase of active synrift tectonism that was associated with the reactivation of extensional faults, whereas during EC4 (late Holkerian to mid-Asbian) and EC6 (early to mid-Brigantian) post-rift periods occurred. These tectonic changes, in combination with sea-level variation, therefore resulted in a more stable and uniform succession within the studied core for the duration of Cycle IV.

By defining a slope depositional setting for the succession (Fig. 5) it is possible to interpret the smaller scale sedimentological features observed, such as mudstone lenses and organic matter variations, within that context.

Origin and significance of mudstone lenses in the slope succession

Clay-rich mudstone lenses, referred to as lenticular fabric, are common within both claystone and mudstone lithofacies within the Marl Hill succession. The lenses range from 50 to 800 µm, with smaller lenses found in Facies 1 (calcareous claystone) compared with Facies 3 (argillaceous mudstone). They appear elongated in shape, have tapering uneven edges with no apparent borders (i.e. no appearance of burrow linings) and show apparent alignment along bedding (Fig. 4b, e and h). Lenticular fabric has been observed in a wide range of mud-rich settings and various processes or sources for this fabric have been proposed, including bioturbation, faecal pellets or re-sedimented shale clasts (e.g. Wignall 1994; Rohl et al. 2001; Schieber et al. 2007; Gabbott et al. 2010; Davies et al. 2012; Ghadeer & Macquaker 2012; Könitzer et al. 2014; Schieber 2016a). The lenses observed within the succession are unlikely to be produced by bioturbation or faecal pellets owing to their elongated profile, the shape and distinction of the lens edges, and alignment with bedding as well as the wide range in size. Instead, the lenses resemble those produced in flume experiments by ‘soft’ water-rich (75%) shale intraclasts (Schieber et al. 2010; Schieber 2016a). Similar features have been identified within other Paleozoic successions from both shelf (Schieber et al. 2007, 2010; Davies et al. 2012) and basin floor successions (Könitzer et al. 2014, 2016; Schieber 2016b).

It has been suggested that the mud intraclasts within the Marl Hill succession were derived from partially lithified, upper to mid-slope sediments, owing to the fine grain size of these intraclasts (McIlreath & James 1978; Stow & Piper 1984) and the complex basin morphology of the Bowland Basin (half-graben within a shallow epicontinental seaway with long-lived palaeogeographical barriers; Fig. 1) reducing the likelihood of long distance transport from either riverine or shelf processes (Fraser & Gawthorpe 1990; Davies 2008; Waters et al. 2009, Woodcock & Strachan 2002). It is important to note that flume experiments conducted by Schieber (2016a) concluded that transport of ‘soft’ mud-rich clasts could occur over distances of the order of thousands of kilometres; therefore, it is not possible to entirely discount contribution from riverine transport or shelf processes in the source of mud intraclasts within the succession.

The presence of mud intraclasts within a mud-rich slope succession, such as Marl Hill, is significant as not only does it provide evidence of traction, even within the most fine-grained facies, but also it demonstrates that the succession is not purely the product of hemipelagic settling of material even at times of lowest sediment accumulation rates on the contemporaneous shelf (Facies Association A; Fig. 5). In addition, it demonstrates that the process of erosion, transport and deposition of soft water-rich mud clasts can occur within a slope environment as well as both shelf and basin floor successions.

Distribution of organic matter in the slope succession

Organic matter (OM) content within the succession is highly variable (1.05 to < 11.16 wt% TOC). Within the succession between 64 and 5 m (Facies Association A, Cycle IV, Bowland Shale Formation; Fig. 3) detailed sampling allows the organic characteristics of single lithofacies to be established. Calcareous claystones (Facies 1) contain the highest organic matter content (average 5.70 wt%; 3.70 to < 11.16 wt.%; n = 6), with Facies 2, calcareous mudstones (average 4.23 wt%; 1.21 to < 5.88 wt%; n = 15), Facies 3, argillaceous mudstone (average 3.81 wt%; 3.30 to < 4.53 wt%; n = 6), Facies 4, calcareous siltstones (average 3.58 wt%; 1.21 to < 4.98 wt%; n = 8) and Facies 5, calcareous sandy mudstone (average 1.57 wt%; n = 1), showing a broad trend of decreasing organic matter content with increasing grain size.

The depositional processes that created each facies were probably instrumental in controlling the nature and content of the facies-specific organic matter (Table 1). The lithofacies with the greatest contribution from hemipelagic settling (Facies 1) show the greatest enrichment of organic matter. RockEval™ data (Figs 6 and 7; Facies 1) suggest kerogen type II–III, indicating a mixed marine and terrestrial source of organic matter. Mixed type II–III kerogen possibly resulted from the clay-rich lenses observed within the facies that contributed a reworked type III kerogen signature to the organic matter from hemipelagic settling (typically) type II kerogen.

The lithofacies primarily deposited as a result of low-density turbidite deposits, Facies 2, 3 and 4, consist of mixed type II–III kerogen. Calcareous mudstones (Facies 2) differ from argillaceous mudstones (Facies 3) in the nature of the organic matter they contain, when observed using back-scattered electron (BSE) SEM (Fig. 4). Argillaceous mudstones contain predominantly fragmentary ‘solid’ OM, inferred to originate from more terrigenous sources (Type III kerogen; Fig. 4i), although calcareous mudstones contain predominantly amorphous OM (OM that aligns along bedding and distorts around grains), inferred to originate from marine sources (Type II kerogen; e.g. Hackley & Cardott 2016; Newport et al. 2016; Fig. 4f). Organic geochemical parameters (i.e. HI, S2) also suggest a contrast in kerogen type between the two facies, where argillaceous mudstone plots close to or within Type III kerogen and calcareous mudstone plot close to or within Type II kerogen (Figs 6 and 7; Table 2). This suggests that sediment and OM for the argillaceous mudstones were derived from a predominantly terrestrial clastic source such as riverine input, and the sediment and OM for the calcareous mudstones were derived from a more marine carbonate source such as shelf carbonate reefs or isolated carbonate build-up (i.e. mudmounds; Fig. 5).

The lithofacies primarily deposited as a result of low-strength, cohesive debrites, calcareous sandy mudstone (Facies 5), contain the lowest TOC content of all facies measured (average 1.57 wt%). RockEval™ data for Facies 5 suggest presence of Type III kerogens with low HI values and relatively high OI values (90 and 13, respectively; Fig. 7; Table 2). These geochemical data suggest a predominantly terrestrial origin for the organic matter within Facies 5, contrasting with the sedimentological evidence, such as carbonate allochems, suggesting an origin from slope instability and collapse of carbonate build-ups (Table 1; Fig. 4). However, the mineral matrix can have a significant influence on the RockEval™ kerogen typing, particularly with sediments consisting of less than 2 wt% organic carbon, such as Facies 5, leading to uncertainty in the data. Hydrocarbon retention on mineral grains can result in a reduction in the HI, and thermal decomposition of small amounts of carbonate during analysis can increase the OI by contributing additional carbon dioxide (Espitalié et al. 1980; Katz 1983; Dembicki 2009). Another explanation for the Type III signature for this facies is a mixture of Type II, III and IV kerogens (Dembicki 2009). Type IV (inertinite) kerogen would contribute to the TOC value but not significantly to the S2 values, therefore lowering HI values reported; however, from Type IV kerogens one would also expect an increase in OI values above that observed (Fig. 6; Table 2). Another consideration when using RockEval™ data to type kerogens present is the maturity of the sediments, as Type I, II and III RockEval™ indices evolve with maturation of the respective hydrocarbons (Dembicki 2009). All samples taken from the Marl Hill succession plot within the oil window of the modified van Krevelen plot, indicating that they are oil-mature, with average Tmax values of 438°C (Fig. 7; Table 2), and thus it is appropriate to use RockEval™ indices to estimate kerogen type.

Within Facies Association A, the minimum TOC value of c. 3.70 wt%, observed for Facies 1, may reflect the background organic matter supply and preservation on the sediment-starved slope setting where the minimum amount of sediment dilution occurs, whereas the minimum TOC value of c. 1 wt% for Facies 4 may reflect the background organic matter supply and preservation on the sediment-starved slope setting where the maximum amount of sediment dilution occurs. Variations can be observed in organic matter content between facies associations; Facies Association A (TOC average 3.95, lowest 1.05 and highest 11.16 wt%) and Facies Association B (TOC average 3.54, lowest 1.21 and highest 9.55 wt%) show a relative enrichment in Facies Association A relative to Facies Association B of the order of 1 wt%. Variation can also be observed in the type of kerogens within the facies associations, with Facies Association B consisting of a higher degree of Type III material than Facies Association A; this may be due to the nature of the prograding shelf contributing a more significant signature from terrestrial organic material (Fig. 5). Variation in the type and quantity of organic matter between facies associations is significant, as it allows the prediction of organic matter content and type to be made based on the relative sea level (i.e. sequence stratigraphic framework) of a slope succession.

Conclusions

  1. Six lithofacies, comprising calcareous claystone, calcareous mudstone, argillaceous mudstone, calcareous siltstone, calcareous sandy mudstone and matrix-supported carbonate conglomerate, are identified in core MHD11. Depositional processes in all are dominated by turbidites and/or debrites. There are three facies associations: a sediment-starved slope, a slope dominated by turbidites and a slope dominated by debrites. Conditions of sediment starvation on the slope occurred during flooding of the contemporaneous shelf, when shorelines and loci of active clastic and carbonate sedimentation were furthest away from the slope. Nevertheless, active sedimentation did occur on the slope at this time, via dilute turbidites, derived from storm-generated or river-effluent-generated density flows that traversed the shelf, or via slumps and slides on the upper slope that transformed into density flows. The transition towards a slope dominated by turbidites and then debrites occurred during normal or forced shoreline progradation towards the shelf margin. Large clasts of carbonate that occur within the slope dominated by debrites facies association are interpreted as the deposits of collapsed shelf-edge carbonates at the time of maximum shoreline progradation.

  2. The Bowland Shale Formation largely consists only of sediment-starved slope sediments, which resulted from an extended period (c. 5 myr) of low quantities of sediment delivery to the slope, with carbonate and clastic sediments trapped on the shelf, owing to a largely flooded shelf.

  3. The presence of mud intraclasts within a mud-rich slope succession provides evidence of traction, even within the most fine-grained facies, rather than the sediment being purely the product of hemipelagic settling. The presence of mud intraclasts also demonstrates that the process of erosion, transport and deposition of soft water-rich mud clasts can occur on a slope, as well as shelves and basin floors, as previously described.

  4. The sediment-starved slope succession is dominated by Type II kerogen, whereas the slope dominated by low-density turbidites is dominated by Type III kerogen. The difference is interpreted to represent a greater contribution from (reworked) deposits of marine hemiplegic settling on the sediment-starved slope, and a greater contribution from terrestrially derived organic material on the slope dominated by turbidites.

  5. This study suggests that mud-dominated lower slope settings are largely active depositional sites, where there is consistent evidence for sediment traction. Additionally, the composition and texture of basin slope mudstones, as well as organic content, vary predictably as a function of shelf processes linked to relative sea-level change.

  6. Variation in composition and texture of slope mudstones, as well as organic matter concentration and type, has been observed within the core studied. This approach can be applied within a basin with a pre-existing sequence stratigraphic framework as a result of greater data availability, and will facilitate predictions of porosity, permeability, mechanical properties, TOC and kerogen type.

  7. In contrast to previous assumptions, this study demonstrates that mud-dominated lower basin slopes are active depositional sites resulting in a wide range of deposits from both turbidity currents and debris flows.

Acknowledgements

The authors thank M. Jones at Newcastle University and G. Hansen at APT Norway for analytical services provided, as well as the British Geological Survey for the loan of material. S.M.N. thanks L. Newport for the many discussions relating to this paper. E.H. publishes with the approval of the Executive Director of the British Geological Survey, Natural Environment Research Council.

Funding

S.M.N. thanks NERC EAO DTP for provision of funding (NE/L002469/1) for this PhD research.

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0/)