Abstract

Sediment transport and distribution are the keys to understanding slope-building processes in mixed carbonate-siliciclastic sediment routing systems. The Permian Bone Spring Formation, Delaware Basin, west Texas, is such a mixed system and has been extensively studied in its distal (basinal) extent but is poorly constrained in its proximal upper-slope segment. Here, we define the stratigraphic architecture of proximal outcrops in Guadalupe Mountains National Park in order to delineate the shelf-slope dynamics of carbonate and siliciclastic sediment distribution and delivery to the basin. Upper-slope deposits are predominantly fine-grained carbonate lithologies, interbedded at various scales with terrigenous (i.e., siliciclastic and clay) hemipelagic and gravity-flow deposits. We identify ten slope-building clinothems varying from terrigenous-rich to carbonate-rich and truncated by slope detachment surfaces that record large-scale mass wasting of the shelf margin. X-ray fluorescence (XRF) data indicate that slope detachment surfaces contain elevated proportions of terrigenous sediment, suggesting that failure is triggered by changes in accommodation or sediment supply at the shelf margin. A well-exposed terrigenous-rich clinothem, identified here as the 1st Bone Spring Sand, provides evidence that carbonate and terrigenous sediments were deposited contemporaneously, suggesting that both autogenic and allogenic processes influenced sediment accumulation. The mixing of lithologies at multiple scales and the prevalence of mass wasting acted as primary controls on the stacking patterns of terrigenous and carbonate lithologies of the Bone Spring Formation, not only on the shelf margin and upper slope, but also in the distal, basinal deposits of the Delaware Basin.

1. INTRODUCTION

The dynamics of continental margin evolution and sediment delivery determine the spatial and temporal distribution of reservoir-forming elements (Saller et al., 1989; Bull et al., 2009; Playton et al., 2010; Janson et al., 2011; Stevenson et al., 2015; Hurd et al., 2016; Playton and Kerans, 2018) that record autogenic and allogenic processes acting on the system (Shanley and McCabe, 1994; Covault et al., 2007; Burgess, 2016; Madof et al., 2016; Romans et al., 2016). The importance of stratigraphic architecture and sediment distribution on continental margin evolution has been documented in both siliciclastic (Kertznus and Kneller, 2009; Sylvester et al., 2012; Salazar et al., 2015; Stevenson et al., 2015; Prather et al., 2017) and carbonate (Bosellini, 1984; Sonnenfeld, 1991; Kerans et al., 1993; Ross et al., 1994; Sarg et al., 1999; Mulder et al., 2012) depositional systems. Clinothems (packages of sediment bounded by sigmoidal surfaces) formed in both siliciclastic and carbonate systems record continental-margin evolution and the variable distribution of lithologies (Rich, 1951; Mitchum et al., 1977; Vail, 1987; Sonnenfeld, 1991; Ross et al., 1994; Sarg et al., 1999; Playton et al., 2010; Salazar et al., 2015). Most studies of clinothems have focused on progradational siliciclastic margins (Mitchum et al., 1977; Vail, 1987; Bull et al., 2009; Kertznus and Kneller, 2009; Sylvester et al., 2012; Salazar et al., 2015; Stevenson et al., 2015; Prather et al., 2017) or steep, reef-rimmed carbonate margins (Bosellini, 1984; Katz et al., 2010; Harman, 2011; Mulder et al., 2012; Jo et al., 2015; Principaud et al., 2015; Playton and Kerans, 2018). Studies of low-relief, mixed carbonate-siliciclastic margins are less well documented (Saller et al., 1989; James et al., 1992; Fitchen, 1997; Grosheny et al., 2015; Tassy et al., 2015), although mixed-system deposits form important petroleum reservoirs and well-preserved archives for paleoenvironmental records (Allen et al., 2013; Tassy et al., 2015; Hurd et al., 2018; Chiarella et al., 2019).

In the Delaware Basin of west Texas, the Leonardian Victorio Peak (shelf facies) and Bone Spring (slope to basin facies) formations record a low-relief, mixed carbonate-siliciclastic depositional system that forms a prolific hydrocarbon system (Allen et al., 2013; Driskill et al., 2018; Schwartz et al., 2018). Studies in the Bone Spring Formation have focused primarily on the basinal deposits that record heterogeneity between siliciclastic and carbonate lithologies and a mixture of turbidite, mass-transport, and hemipelagic-pelagic deposits (Saller et al., 1989; Montgomery, 1997a, 1997b; Asmus and Grammer, 2013; Nance and Rowe, 2015; Driskill et al., 2018). A few studies (Kirkby, 1982; Fitchen, 1997) have focused on the shelf (Victorio Peak) deposits, documenting cyclic deposition of platform carbonates and bypass of terrigenous sediment. While the platform (proximal) and basinal (distal) portions of the Bone Spring sediment routing system have been well documented, the upper-slope segment that is important for sediment transfer to the deep ocean is only partially documented (King, 1948; McDaniel and Pray, 1967; Kirkby, 1982; Fitchen, 1997).

This study constrains the progradational slope architecture and sediment distribution of the upper-slope Bone Spring deposits exposed in Guadalupe Mountains National Park, west Texas. We document (1) slope-building clinothems of variable and mixed lithology, (2) slope detachment surfaces bounding clinothems, and (3) abundant sediment gravity flow deposits and their genetic relationships to clinothems and slope detachment surfaces. These observations provide the basis for discussion of slope evolution on a mixed-lithology margin, the role of mass wasting and terrigenous sediment supply in shaping the margin, delivery of sediment to the basin via allogenic and autogenic forcing of sediment delivery, and how the stratigraphic evolution of the upper slope affects depositional processes and stacking patterns of carbonate and siliciclastic sediment in the distal Delaware Basin.

2. GEOLOGIC AND STRATIGRAPHIC SETTING

2.1 Geologic Setting

The Bone Spring Formation was deposited in the Delaware Basin, a sub-basin of the larger Permian Basin of west Texas during Leonardian time (middle Permian, ca. 275–280 Ma; Fig. 1). During the late Mississippian assembly of the supercontinent Pangea (ca. 326 Ma), the Permian Basin formed as a foreland basin north of the Marathon-Ouachita-Sonora orogeny (Poole et al., 2005; Fig. 1A inset). Compression reactivated Precambrian areas of weakness and uplift of the Central Basin Platform, creating two sub-basins—the Delaware and Midland Basins (Fig. 1A inset; Hills, 1984; Hill, 1996; Amerman, 2009; Nance and Rowe, 2015). The Delaware Basin was bounded to the west and north by the Diablo Platform and Northwest Shelf, to the south by the Marathon-Ouachita-Sonora fold belt and Hovey Channel, and to the east by the Central Basin Platform and San Simon and Sheffield Channels (Fig. 1A inset; Asmus and Grammer, 2013). Tectonic activity occurred until at least the middle Wolfcampian (ca. 295 Ma; Hills, 1984; Amerman, 2009), but the Middle Permian Leonardian stage (ca. 285–275 Ma) was generally tectonically quiescent (Hills, 1984; Amerman, 2009). Subsidence related to sediment loading and isostatic adjustment on the basin margins created a deep (~450 m) basin with up to 2500 m of Permian sediment (Hills, 1984).

Carbonate factories on the margin were prolific throughout the Permian and account for large contributions of sediment (Kirkby, 1982; Hills, 1984; Harman, 2011). Carbonate production in the early Permian was dominated by packstone- and grainstone-bank margins of reworked skeletal debris (McDaniel and Pray, 1967; Kirkby, 1982) and transition to boundstone- and rudstone-reef–rimmed margins in the late Permian (Hills, 1984; Harman, 2011). While carbonate sediment production occurred all around the basin margins, terrigenous (i.e., siliciclastic) sediment entering the Delaware Basin was predominantly sourced from the north and east where aeolian and fluvial sediments were deposited on the shelf and shelf margin (Presley, 1987; Fischer and Sarnthein, 1988; Soreghan and Soreghan, 2013), with some terrigenous input from the Marathon-Ouachita-Sonora region to the south (Soto-Kerans et al., 2020). During Leonardian time, and especially during low sea-level conditions, the entrance to the Panthalassa Ocean to the west was restricted by a sill in the Hovey Channel (Fitchen, 1997); this sill hindered water circulation in the basin, resulting in euxinic conditions (McDaniel and Pray, 1967), minimal bioturbation, and preservation of organic-rich sediment (Hills, 1984).

2.2 Shelf-to-Basin Stratigraphy

The evolution of the shelf-margin and basinal strata of the Delaware Basin is well documented (King 1948; Sarg and Lehmann, 1986; Kerans et al., 1993; Sarg et al., 1999; Kerans and Kempter, 2002). Figure 2 shows the correlation of the chronostratigraphic and lithostratigraphic units from shelf to basin. The Wolfcamp Formation (the lowermost Permian deposits) is a mixed carbonate-siliciclastic, prograding, shelf-to-basin system (Silver and Todd 1969; Kvale and Rahman, 2016) but does not outcrop in the study area (Fig. 1B). It is overlain by Leonardian prograding carbonate banks to rimmed platforms with 5°–20° slopes (Harris, 2000) that transition into a deep basin assemblage (Fig. 2; Fitchen, 1997; Asmus and Grammer, 2013; Hurd et al., 2018). The Leonardian system is composed of the proximal Yeso Formation, which represents a restricted shelf environment with aeolian red beds and evaporitic deposits (Stanesco 1991; Fitchen, 1997). The Yeso Formation transitions to the bank-rimmed carbonate grain margin of the Victorio Peak Formation (Kirkby, 1982); this margin transitions to the Bone Spring Formation carbonate and siliciclastic slope and basin deposits (Fig. 2; Saller et al., 1989; Fitchen, 1997; Montgomery, 1997a; see Fig. 2). Fitchen (1997) described six third-order sequences in the Victorio Peak–Bone Spring margin (L1–L6; Fig. 2), in which sequence boundaries reflect subaerial exposure of the carbonate platform and coeval siliciclastic deposition in the deep basin. Within sequences, lowstand (siliciclastic-rich) and highstand to transgressive (carbonate-rich) members are thought to reflect cyclicity in sea-level and basin subsidence (Fig. 2; Silver and Todd, 1969; Saller et al., 1989; Fitchen, 1997; Nance and Rowe, 2015). Higher-order cyclicity within both siliciclastic and carbonate-rich members (Montgomery, 1997a; Nance and Rowe, 2015; Driskill et al., 2018) is interpreted as reflecting high-frequency variations driven by allogenic sea-level forcing (Nance and Rowe, 2015). A significant erosional surface separates the Victorio Peak and Bone Spring Formations from the overlying Cutoff Formation (Fig. 2), variously referred to as the top of LD10 (Sarg et al., 1999) or top of L6 (Fitchen, 1997; Hurd et al., 2018). Early deposition of the Cutoff Formation reflects a lowstand system that eroded parts of the Victorio Peak–Bone Spring margin before reaching a maximum transgression (L8/G1) biostratigraphically correlated with the Leonardian to Guadalupian boundary (Hurd, 2016; Hurd et al., 2016). Overlying the Cutoff Formation is the Guadalupian Delaware Mountain Group, including the Brushy Canyon Formation (G5-G7), consisting of a submarine channel-fan system (Fig. 2; Zelt and Rossen, 1995; Gardner and Sonnenfeld, 1996; Gardner et al., 2008). Capping the succession are Guadalupian reef-rimmed carbonate platforms (Capitan Formation) and their coeval basinal deposits (Fig. 2; Kerans et al., 1993; Harman, 2011).

It is difficult to constrain the exact age of Victorio Peak and Bone Spring outcrops in Guadalupe Mountains National Park (Fig. 1) because of paleo-erosional features (e.g., Cutoff Formation) and poor-resolution biostratigraphy. Lithostratigraphic correlations from Fitchen (1997) suggest that the Victorio Peak–Bone Spring outcrops represent the L5 and L6 shelf margin to upper-slope sequences, and this interpretation is supported by recent biostratigraphic and lithostratigraphic correlations in the Cutoff Formation (Hurd et al., 2016). However, to complicate matters, correlation of the Bone Spring Formation outcrops into the subsurface is difficult, because many industry naming schemes are purely lithostratigraphic (e.g., 1st Bone Spring Sand, 1st Bone Spring Carbonate, Avalon Shale) and absolute age control (e.g., biostratigraphy) is lacking (Fig. 2; Driskill et al., 2018). Hurd et al. (2018) correlate the base of the Cutoff Formation outcrops (base L7) to the base of the Upper Avalon Shale in the basin (Fig. 2). On this basis, outcrops of the Bone Spring Formation in the study area likely correlate to basinal rocks referred to as the Middle Avalon Carbonate, Lower Avalon Shale (boundary between L5 and L6), and some portion of the 1st Bone Spring Carbonate and 1st Bone Spring Sand (Fig. 2).

3. STUDY AREA AND OUTCROP MAPPING

3.1 Study Area

The study area lies along the “western escarpment” of Guadalupe Mountains National Park (Fig. 1B), a northward-trending footwall fault block created during Cenozoic extensional tectonism (Hills, 1984; Hill, 1996) that exposes Leonardian and Guadalupian carbonate and siliciclastic shelf-margin stratigraphy (Fig. 2; King, 1948; Hills, 1984; Harris, 1987). Postdepositional loading (Hills, 1984), Late Cretaceous transpression (Montgomery, 1997a), and the growth of the Cenozoic Huapache Monocline (Hayes, 1964; Resor and Flodin, 2010) contribute to a 2°–4° eastward dip of Permian rocks along the escarpment. King (1948) extensively mapped the area, including the Bone Spring Formation, which is well exposed in a system of west-east–trending canyons (Fig. 1B). We focus on outcrops in Shumard and Bone Canyons, and the west-facing exposures linking the canyons (Fig. 1B). The historic Williams Ranch House (Fig. 1B), built in 1908, stands at the entrance to Bone Canyon.

3.2 Three-Dimensional Outcrop Model and Field Measurements

A three-dimensional digital outcrop model was built using Agisoft software and >2000 drone-derived photographs (Fig. 3). Using the existing stratigraphic framework, the study area was constrained below the Cutoff Formation and down-dip from the lithostratigraphic boundary with the Lower Victorio Peak Formation (Fig. 3). Field observations from bedding-attitude transects (N = 16 transects; n = 593 bedding measurements), nine measured sections, and six photopanel interpretations were incorporated into the model to capture facies relationships, depositional elements, and prominent stratigraphic surfaces (Fig. 3).

4. LITHOFACIES AND DEPOSITIONAL ENVIRONMENTS

4.1 Lithofacies

Lithofacies naming schemes can be difficult in mixed carbonate-siliciclastic systems because of differences between schemes based on texture and/or composition (e.g., Dunham, 1962; Folk, 1980) and those based on interpretations of depositional process and stratigraphic architecture (e.g., Bouma, 1962; Lowe, 1982; Hubbard et al., 2008). This is especially true when tying in descriptions from a carbonate platform to carbonate-siliciclastic basinal deposits. For the purposes of this paper, we use a system-scale lithofacies scheme based on the historical naming convention with the highest constituent component (Fig. 4C). Thus, if carbonates form >50% of the sediment, we use the Dunham classification (Dunham, 1962), and if siliciclastic sediment is >50% of the sediment, we use the Folk classification (Folk, 1954, 1980). To further clarify the composition of the lithofacies, we add a modifier if a secondary constituent makes up greater than 10% of the sediment (e.g., bioclastic quartz siltstone; Chiarella and Longhitano, 2012; Lazar et al., 2015). Sedimentary structures are also added as a modifier to capture distinguishing characteristics between lithofacies. Further work is needed to develop a unified naming scheme in these types of systems (see efforts from Chiarella and Longhitano, 2012, and Lazar et al., 2015).

4.1.1 Descriptions

Eight lithofacies are identified based on composition and/or lithology, grain size, bed thickness, sedimentary structures, and fossil content (Figs. 4 and 5; Table 1). In addition to field observations, lithofacies were constrained by thin section analysis, scanning electron microscope (SEM) analysis, and X-ray fluorescence (XRF) analysis. The eight lithofacies listed are based on observations outlined in Table 1 and Figure 4: (F1) thin-bedded laminated lime mudstone; (F2) thin- to thick-bedded deformed lime mudstone; (F3) thick-bedded bioclastic lime wackestone to packstone; (F4) interbedded lime mudstone and bioclastic packstone; (F5) thick-bedded normally graded bioclastic lime packstone to grainstone; (F6) thin-bedded laminated bioclast quartz siltstone; (F7) thin-bedded laminated quartz lime mudstone; and (F8) thick-bedded bioclastic lime packstone to grainstone.

4.1.2 XRF Relationships and Compositional Mixing

The partitioning and mixing of sediments on the Bone Spring slope are constrained by handheld XRF measurements (Fig. 4B). Typically, Ca versus Si plots distinguish carbonate from siliceous clay and siliciclastic sediment, but biogenic silica is abundant in the Bone Spring Formation due to sponge spicules and radiolaria that have been diagenetically altered to chert (Figs. 5A, 5B, 5C, and 5H; McDaniel and Pray, 1967). We use “terrigenous” to indicate clay and siliciclastic sediment delivered from a non-biogenic source (i.e., onshore) and “biogenic silica” to indicate silica produced biotically. Petrography in the Bone Spring Formation (Table 1) shows the primary terrigenous material present is composed of quartz silt and sand and clay minerals (e.g., illite). However, XRF data do not distinguish biogenic silica from terrigenous silica, and thus researchers typically choose a terrigenous sediment proxy (e.g., Si/Al, Zr/Al, and Zr/Cr ratios in Driskill et al., 2018). Here we choose Al + Ti (cf. Tribovillard et al., 2006) to create a ternary diagram (Si, Ca, and Ti + Al) that establishes carbonate and terrigenous domains (Fig. 4B). The trend line in Figure 4B represents the continuum of carbonate-terrigenous compositional mixing, and vertical deviation from this trend line suggests the presence of biogenic silica. Samples that plot significantly below the trend line in Figure 4B have high silica, but little to no terrigenous sediment; for example, some samples have high Si but low Ti + Al (e.g., two blue dots of Lithofacies 5 highlighted in Fig. 4B). Thin-section analysis (Fig. 5E) reveals that these samples contain little to no terrigenous sediment and are cemented by silica of biogenic origin (i.e., chert) that is not derived from any terrigenous source. The most common lithofacies in the study area are carbonate mudstones to packstones composed predominantly of carbonate mud and skeletal grains (F1–F5) that plot in the carbonate domain, but with variable compositional mixing with a terrigenous input (Fig. 4B). Thin sections reveal little to no terrigenous clay and a small volume (<10%) of well-rounded, silt-sized, quartz grains (Figs. 5A5D) that are interpreted as aeolian dust transported from onshore ergs (Presley, 1987; Fischer and Sarnthein, 1988; Cecil et al., 2018). The F6 siltstone lies within the terrigenous domain (Fig. 4B) along the trend line and, thus, are not influenced by chert. Thin sections (Fig. 5F) reveal F6 is composed predominantly of silt-sized quartz and siliceous clays with minor (<25%) carbonate skeletal grains (Table 1). F7 shows varied compositional mixing of terrigenous and carbonate sediments (Fig. 4B; Chiarella et al., 2017) and these mixed facies (F7) plot along a continuum between carbonate and terrigenous domains.

4.1.3 Interpretations

The depositional processes forming each lithofacies are interpreted from descriptive observations outlined in Table 1 and Figure 4. F1 is interpreted as hemipelagic and sediment gravity-flow deposits (Fig. 5A; Schieber et al., 2007; Talling et al., 2012; Birgenheier and Moore, 2018). The compositional mixing of this lithofacies suggests a carbonate-dominated environment with minor influx of aeolian quartz. F2 comprises mass-transport deposits (MTDs) (Fig. 5B; Jablonská et al., 2018) originally deposited as F1, but remobilized and deformed. F3 is interpreted as more proximal hemipelagic and sediment gravity-flow deposits relative to F1 (Fig. 5C). The thin packstones of F4 and the thick-bedded packstones and grainstones of F5 are interpreted as turbidity current deposits (Bouma, 1962; Lowe, 1982) transporting coarse-grained platform sediment downslope (Figs. 5D and 5E). F4 deposits represent low-density flows interbedded with F1 mudstones, whereas the amalgamated F5 deposits are high-energy turbidites. F6 facies are composed of deposits of siliciclastic silt and terrigenous clays (75%) containing only minor carbonate skeletal fragments, suggesting deposition by hemipelagic and sediment gravity flows transported over a carbonate shelf (Fig. 5F). F7 is interpreted as of similar origin to F6, but with higher compositional mixing with carbonate skeletal fragments, possibly from minor contemporaneous carbonate production (Fig. 5G). Finally, F8 is interpreted as carbonate platform deposits either formed in-place or reworked (Fig. 5H); these deposits represent the Lower Victorio Peak of King (1948) and Kirkby (1982).

4.2 Depositional Environments

Four facies associations in outcrops of the Bone Spring Formation are interpreted to represent sub-environments within the Bone Spring mixed-lithology, shelf-slope depositional system (Figs. 6 and 7; see Fig. S3 for location of figures).

4.2.1 Facies Association 1

Description. Facies Association 1 (FA1) consists of F1, F3, and F8, with a predictable stacking pattern of F1 at the base, F3 in the middle, and F8 at the top (Fig. 6A) with decreasing chert content and increasing grain size, bed thickness, and macro-fossil content from Lithofacies 1 to Lithofacies 8 (Fig. 4A). Typical thickness for an FA1 deposit is ~30 m, with gradational contacts between lithofacies (Fig. 6A).

Interpretation. FA1 represents an upward-shoaling sequence from slope carbonate mudstones to platform carbonates (Figs. 6A and 7A; McDaniel and Pray, 1967). Planar and ripple lamination in F1 suggest that low-density muddy turbidites were a dominant process in slope development of the Bone Spring Formation (Figs. 5A and 8; cf. Birgenheier and Moore, 2018). The upward-shoaling character supports previous studies suggesting that the Leonardian carbonate margin was progradational in the study area (Kirkby, 1982; Fitchen, 1997; Sarg et al., 1999). The facies transition of FA1 is best exposed in Shumard Canyon, where steeply dipping (~15°) Bone Spring slope deposits are overlain by and transition northwestwards into flat-lying Victorio Peak Formation (F8), which forms the uppermost cliffs (Fig. 3B). Bedding attitude data show that the shelf-slope system built out in a predominantly eastward direction but varied in orientation from 060° to 180° (Fig. 3A).

4.2.2 Facies Association 2

Description. FA2, the primary facies association on the outcrop, makes up roughly 90% of the study area and includes interbedded mixtures of F1, F2, and F4 (Figs. 6B and 7B), commonly transitioning laterally between lithofacies. Contacts between lithofacies are predominantly sharp, with truncation beneath and onlap above surfaces, particularly between F1 and F2 (Figs. 6B and 7B). Most commonly, there is an absence of coarse-grained material directly mantling surfaces; however, surfaces are often draped by fine-grained sediment. Chaotic bedding and synsedimentary folds observed in F2 (Figs. 5B and 7B) vary in scale and style (Figs. 8 and 9), including creep, slump, and debrite deposits. Creep deposits (sensu Auchter et al., 2016) are observed at many scales (Figs. 8A8C) but are most commonly micro-scale, typically deforming laminae (<1 cm) in brittle (e.g., micro-faulting) and ductile (micro-folding) fashion (Figs. 8B1 and B2). Slump deposits in the study area are composed of carbonate (F1, F2, and F3) or terrigenous (F6) lithofacies, where bedding is generally preserved but plastically deformed (Fig. 8C). Meter-scale slump deposits are most common but can be as thick as 10 m (Fig. 8C). Slump deposits often display a basal shear zone with fracturing and brecciation (sensu Cardona et al., 2020). Debris-flow deposits of carbonate mudstones (Fig. 5B) have minimal preserved strata, chaotic fabric with matrix-supported clasts, brittle deformation features (breccia, fractures), and erosional bases (Fig. 9), features common to debrites (Dott, 1963; Fisher, 1983; Stow, 1984; Moscardelli and Wood, 2008; Tripsanas et al., 2008; Talling et al., 2012).

Interpretation. FA2 represents an unstable carbonate slope with abundant mass failure. The prevalence of mass wasting suggests that the Bone Spring slope was almost always over-steepened and prone to failure (Stow, 1986). Sharp erosional surfaces are interpreted as slope failure scarps rather than erosional bypass surfaces due to the lack of coarse-grained material mantling the surfaces (Fig. 7B). Surface depressions are commonly filled with a wedge geometry that is interpreted as filling of local topography by sediment gravity flows. Similar “failure-and-fill” architecture has been documented in other carbonate slope deposits (Bosellini, 1984; Ross et al., 1994; Katz et al., 2010; Playton et al., 2010; Mulder et al., 2012; Playton and Kerans, 2018).

4.2.3 Facies Association 3

Description. Facies Association 3 (FA3) consists of F1, F3, F5, and F8 (Figs. 6C, 7C, and 7D). The type locale is on the south wall of Shumard Canyon (Fig. 3B), where a sharp surface 100 m wide with 10 m relief truncates F3, with a lenticular deposit of F5 onlapping the surface (Fig. 7C). A 10 m interval of amalgamated F5 gradually transitions upward into F8. Other occurrences of FA3 (Figs. 3B, 6C, and 7D) show similar architecture but smaller dimensions (e.g., 10 m wide and 0.5 m thick, Fig. 7D) and contact of F1 overlain by F5.

Interpretation. FA3 is interpreted as submarine channel deposits developed on a carbonate slope. The erosional truncation of fine-grained lithofacies (F1 and F3) and overlying coarse-grained channel fill with normally graded F5 beds indicates erosion and deposition by turbidity currents (Figs. 5E, 7C, and 7D; Talling et al., 2012; Janocko et al., 2013). Amalgamation surfaces within F5 (Figs. 5E and 7C) suggest that the channels were long-lived conduits for transport of carbonate sediment to the basin. The presence of F8 (Lower Victorio Peak) overlying the channel fill in the type locale suggests that this channel was located very near the shelf edge. Smaller channel deposits in contact with F1 (e.g., Fig. 7D) are interpreted to have been in mid-slope positions and may represent slope gully deposits (Shumaker et al., 2016).

4.2.4 Facies Association 4

Description. Facies Association 4 (FA4) comprises interbedded terrigenous F6 and mixed-lithology F7 deposits (Figs. 6D and 7E). Bed-scale alternations of F6 and F7 typically occur at the ~10 cm scale (Fig. 7E), but on the west wall of Shumard Canyon, F6 deposits are ~10 m thick (Fig. 3B). Contacts between F6 and F7 are typically sharp and undulatory (Fig. 7E). Like FA2, FA4 contains internal truncation surfaces with overlying deformed intervals.

Interpretation. FA4 is interpreted as periods when carbonate and terrigenous sediment were deposited contemporaneously on the Bone Spring slope. Alternations in the proportion of terrigenous sediment suggest fluctuations in carbonate-terrigenous sediment delivery. Deformed intervals and truncation surfaces indicate an unstable slope dominated by failure and bypass, similar to that of FA2 carbonate deposits.

5. STRATIGRAPHIC ARCHITECTURE AND SEDIMENT DISTRIBUTION

5.1 Slope Detachment Surfaces

Six photopanels document the architecture of the Bone Spring Formation within the study area (Figs. 1B and 1014, and Figs. S1 and S2 in the Supplemental Material1). The major architectural features are large-scale (>20 m relief) truncation surfaces (Figs. 3B, 15A, and 15B) that can be mapped the length of the outcrop along-strike to the paleo-shelf (i.e., kilometer-scale; Fig. 3B; Sarg et al., 1999) before disappearing into the subsurface, coalescing with another surface, or transitioning northwestward into the Victorio Peak shelfal lithofacies (black surfaces, blue numbers in Figs. 3B and 10–15). Because these steeply dipping surfaces truncate Bone Spring slope deposits (Figs. 9A and 10), we interpret them as slope detachment surfaces (SDSs) related to large-scale mass wasting of the margin. Slope detachment surfaces are differentiated from smaller-scale discordant surfaces by the amount of truncation (using a cut-off of 20 m). An example of the scale of surfaces can be found toward the back of Bone Canyon where a large-scale surface (SDS 9) is overlain by smaller-scale discordant surfaces (Fig. 15C). There are nine SDSs bounding ten clinothems; clinothems are defined as strata bound by SDSs that are mappable across the study area (Fig. 3B). The SDSs and clinothems are marked by blue and orange numbers, respectively, in Figures 3B and 1014. Clinothems 1–3 and 5–10 are dominated by carbonate deposits (FA2), whereas Clinothem 4 is dominated by terrigenous deposits (FA4; Fig. 3B).

Slope detachment surfaces were identified using bedding attitude changes and truncation/onlap relationships (Figs. 3 and 1015). Slope detachment surfaces typically dip eastward at ~20° (but can be as steep as 45°) and the stratigraphic relief ranges from 20 to 100 m (Figs. 1014). Slope detachment surfaces 3–8 in Figure 10 show each surface truncating bedding with variable attitudes below and above the surface. For example, bedding shifts from 18°/090° (dip magnitude/dip azimuth) below SDS 3 to 23°/045° above it (Fig. 16A), and across SDS 8, a 40° bedding azimuth change is observed (Fig. 16D). Typically, SDSs are underlain by F1 (undisturbed lime mudstone deposits) and overlain by F2 and F4 sediment-gravity-flow deposits (Figs. 16B and 16E). Some SDSs show sigmoidal geometries that flatten and become conformable within the Victorio Peak (SDS 4, 7, 8, 9; Fig. 12; cf. Rich, 1951), suggesting that SDS may be associated with large-scale clinoforms, much like those mapped in the Leonardian (Sarg, 1988; Fitchen, 1997) and Guadalupian (e.g., Harman, 2011) shelf margins.

We interpret slope detachment surfaces as representing evacuation scars of subaqueous mass failure of the margin. Relief along the SDSs suggests that failures were quite large (tens to hundreds of meters of sediment height), consistent with observations of large MTD deposits from the basinal Permian Basin (Saller et al., 1989; Allen et al., 2013; Bhatnagar et al., 2018). The lack of large-scale, thick (>20 m) MTDs in the study area suggests that most were sourced from the steep (~15°) upper slope, bypassing the mid-slope, to be deposited distally at the toe-of-slope or farther into the basin. Similar slope segmentation with large-scale MTDs has been documented in both the Permian Basin (Saller et al., 1989; Montgomery, 1997a; Allen et al., 2013; Nance and Rowe, 2015; Bhatnagar et al., 2018; Hurd et al., 2018, Schwartz et al., 2018) and in other carbonate slope systems (De Blasio et al., 2005; Moscardelli and Wood, 2008; Mazzanti and De Blasio, 2010; Janson et al., 2011; Mulder et al., 2012; Dakin et al., 2013; Principaud et al., 2015; Cardona et al., 2016; Moscardelli and Wood, 2016). Headwall scarps from sediment evacuation on carbonate slopes are commonly steep (~15°; Mulder et al., 2012; Jo et al., 2015; Principaud et al., 2015), consistent with angles observed in the study area (e.g., Figs. 15 and 16). Bedding attitude changes across surfaces, and together with MTDs, submarine channel deposits, and wedge-fill architecture above surfaces (Fig. 15A) suggest that failure scarps were likely scallop-shaped and actively filled by sediment gravity flows.

5.2 Clinothems 1–3

Clinothems 1–3 are dominated by carbonate facies (FA2), and typically basal deposits show disrupted bedding and MTDs (F2), packstone beds (F4), and turbidites (FA3) overlying SDS, and an upward transition to continuous, planar bedding (F1) that is top-truncated by another SDS (Figs. 1014). SDSs are often draped with thin beds (<5 cm, Figs. 16C and 16D) rich in fine-grained terrigenous sediment. Large MTDs can be found above SDS 1 on the west wall of Shumard Canyon (Fig. 12) and along the Shumard trail (Fig. 9). Clinothems 1–3 prograde approximately eastward (Fig. 3A), and the upslope transition to the Victorio Peak is well exposed (Figs. 3B and 12). A Cenozoic normal fault drops Clinothems 1–3 into the subsurface at the entrance to Bone Canyon (Fig. 3B).

5.3 Clinothem 4

Terrigenous-rich deposits (FA4) are well developed only in Clinothem 4 (Fig. 3B). This interval consists of ~20 m of terrigenous-rich deposits in Shumard Canyon (Figs. 10, 12, and 16A) and become progressively thinner and more carbonate-rich southward (Figs. 3B and 13). South of Shumard Canyon, the terrigenous component is discontinuous and is concentrated in local accumulations along the mapped extent of Clinothem 4 (Fig. 3B), but poor exposure hampers the understanding of detailed facies transitions. Slumps are common in the terrigenous-rich beds, but small-scale truncation surfaces are not apparent. The FA4 in Clinothem 4 are truncated by SDS 4, a prominent truncation surface (Fig. 16A), before transitioning into carbonate-dominant sediments in Clinothem 5. Unlike SDSs that bound Clinothems 1–3, the SDS 4 surface has no draping terrigenous-rich interval (Fig. 16A).

5.4 Clinothems 5–10

Clinothems 5–10 are dominated by carbonate-rich lithofacies (FA2) with numerous small-scale truncation surfaces and MTDs, as well as localized FA3 deposits. The depositional motif of Clinothems 5–10 is quite similar to that of Clinothems 1–3, with a thin (<5 cm) terrigenous interval commonly draping the SDS (Fig. 16), overlain by deformed bedding and coarse-grained deposits, which transition upward into undeformed bedding. This motif is visible in Clinothem 8 between SDS 7 and 8 (Fig. 14). Two 5–10-m-thick MTDs mark the base of SDS 6 and SDS 7 and are well exposed and easily accessible along the Shumard Trail (Fig. 9). Regional observations (Kirkby, 1982) and bedding attitude data show that Clinothems 5–7 built out in a predominantly eastward direction similar to Clinothems 1–4, but a prominent southward shift in slope propagation occurs in Clinothem 8, which is well exposed in Shumard Canyon (Fig. 10B). We interpret this variation as recording a local slope inflection point, where a re-entrant focused deposition, generating a high density of slope failure surfaces and MTDs. Four FA3 submarine channel deposits are vertically stacked here (Fig. 10), suggesting that the topographic lows created by failures acted as conduits for coarse-grained sediment gravity flows. This shift in slope propagation direction suggests that the Bone Spring margin prograded as a series of lobate clinothems, supporting previous studies in the basin (Saller et al., 1989; Sonnenfeld, 1991). Clinothems 5–10 are variably top-truncated by the Cutoff and Brushy Canyon formations (Fig. 3B); notably, the Cutoff Formation contains large rafted blocks of F8 (Victorio Peak lithofacies) that truncate and deform Clinothem 9 on the south wall of Bone Canyon (Fig. 14; cf. Hurd et al., 2016).

5.5 Terrigenous and Carbonate Sediment Distribution

To understand the role of terrigenous and carbonate sediment contributions to the development of slope detachment surfaces, XRF transects were collected across SDS 4, 6, 7, 8, and 9 (Fig. 17) from 1 m below the surface to 1 m above the surface at 20 cm intervals, including five samples taken along the surface itself (Fig. 17). Slope detachment surfaces are enriched in terrigenous sediment (Si*[Ti + Al]) relative to carbonate sediment (Ca) (Fig. 17). The carbonate, mixed, and terrigenous domains in Figure 17 are calculated using the trend line in Figure 4B, with Si*(Al + Ti) as the proxy to detect the presence of terrigenous sediment from XRF data. Each SDS is enriched in terrigenous sediment near the surface relative to the samples taken from within the clinothems (i.e., not adjacent to SDS; cf. Figs. 4B and 17) and, with one exception, begin within the mixed or carbonate domain below the surface and shift toward the terrigenous domain at or near the SDS before reverting to the mixed or carbonate domain above the SDS (Figs. 17B17E). The exception to this trend is SDS 4 (Fig. 17A), in which all XRF data (below, on, and above the surface) lie within the terrigenous domain; it is SDS 4 that is associated with FA4 deposits in Clinothem 4 (Figs. 3B and 16A).

Terrigenous sediment appears to be associated with slope detachment surfaces (i.e., failures), thus we envision three possible mechanisms contributing to failure: (1) loading from increased terrigenous sediment supply (Sultan et al., 2004; Vanneste et al., 2014), (2) weakened substrate from increased terrigenous (i.e., clay) input (Kenter and Schlager, 1989; Kenter, 1990; De Blasio et al., 2006), (3) steep relict slopes created by the carbonate shelf-margin (Schlager and Camber, 1986; Ross et al., 1994). A combination of these may have initiated large-scale slope failure, and the XRF data (Figs. 17B17E) suggest that only a slight increase in terrigenous sediment was necessary to trigger large-scale slope failure. For example, the steep Bone Spring slope (10°–20° non-decompacted) already exceeds the predicted stability limits for muddy carbonate margins (Kenter, 1990); while a change in boundary conditions would not be necessary to initiate failure (e.g., small-scale failure surfaces in Fig. 15), any disturbance to the margin (e.g., relative sea level) coupled with terrigenous sediment delivery may have promoted larger, more widespread failure and creation of a mappable slope detachment surface (SDS).

6. DISCUSSION

6.1 Evolution of the Victorio Peak–Bone Spring Mixed Margin

Sediment mixing and partitioning is a well-documented primary control on slope evolution and architecture (e.g., Bosellini, 1984; Gómez-Pérez et al., 1999; Eggenhuisen et al., 2010; Hurd, 2016). Slope detachment surfaces of the Bone Spring represent large volumes of missing rock on the slope. MTDs associated with these evacuation surfaces either initiated in this area or bypassed it, and were likely deposited at the toe-of-slope and in the basin (Saller et al., 1989; Montgomery, 1997a; Allen et al., 2013; Nance and Rowe, 2015; Bhatnagar et al., 2018; Hurd et al., 2018; Schwartz et al., 2018). Additionally, flow transformation (Fisher, 1983; Haughton et al., 2009; Talling et al., 2012) of MTDs along the sediment routing system may be reflected in the abundance of hybrid event beds documented in the distal Delaware basin (Driskill et al., 2018; Kvale et al., 2020). Terrigenous deposits draping the SDS (Figs. 16C and 16D) may represent bypass surfaces that are coeval with terrigenous and mixed-lithology basinal deposits (Montgomery, 1997a; Asmus and Grammer, 2013; Nance and Rowe, 2015), particularly when compared to the rest of the slope deposits that are predominantly carbonate-rich.

The characteristics of SDS and clinothems, coupled with lithofacies distributions and regional observations (Saller et al., 1989; Montgomery, 1997a; Asmus and Grammer, 2013; Nance and Rowe, 2015), allow us to reconstruct the local Leonardian sediment routing system (Fig. 18) to explore controls on slope-building processes and sediment delivery on a mixed-lithology margin. Four possible evolutionary steps are detailed (A, B, C, and D), and the specific geometries created may vary both laterally and temporally due to along-strike variability inherent in shelf-slope margins (Fig. 10; Saller et al., 1989; Madof et al., 2016; Chiarella et al., 2019). We suggest that this reconstruction and associated principles can be applied to other parts of the basin and in similar mixed-lithology ancient and modern systems.

Clinothem packages 1–3 and 5–10 (Fig. 3B) represent the stratigraphic record of time step A. During this period (Fig. 18A), accommodation-to-sedimentation (A/S) is high (i.e., A/S>1; Shanley and McCabe, 1994), promoting high carbonate production with minimal terrigenous input. Calcareous hemipelagic and sediment gravity flow deposits (FA1 and FA2) dominate the slope and basin, with perhaps limited aeolian contribution (Fig. 5A; Cecil et al., 2018). The slope builds out with spatially variable progradation and aggradation, accounting for temporal changes in carbonate production and along-strike variability in slope morphology (Fig. 10; Saller et al., 1989). The dominance of carbonate lithofacies creates a relatively stable, albeit steep (~15°), slope characterized by local intrastratal deformation and small-scale slope-attached MTDs (Fig. 18A; Moscardelli and Wood, 2008).

Time Step B is represented by SDS 1, 2, and 5–9 (Fig. 18B). Terrigenous sediment supply increases, driving a decrease in A/S (e.g., A/S approaching 1) and destabilizing the shelf-margin and upper slope. This results in large, shelf-attached failures and associated slope detachment surfaces. The SDS may form part of a larger clinoform surface that is traceable onto the shelf and into the basin (Figs. 12 and 18; Sarg, 1988; Fitchen, 1997). During this time, terrigenous sediment largely bypasses the slope but leaves the surface of the SDS relatively enriched in terrigenous material (Figs. 17 and 18E; see also Armitage et al., 2009; Amerman et al., 2011; Grosheny et al., 2015; Stevenson et al., 2015).

In time step C, represented by SDS 3 (Fig. 18C) and Clinothem 4 (Fig. 3B), further decrease of A/S (approaching 0 or negative) introduces larger volumes of terrigenous sediment to the shelf edge and slope. The resulting clinothem is built by FA4 deposits, with the amount of terrigenous (Lithofacies 6) and mixed (Lithofacies 7) sediment dependent on local sediment distribution (Fig. 18C). The steep, inherited slope promotes the bypass of terrigenous sediment into the basin (Fig. 18E).

In time step D (Fig. 18D) A/S returns to time step A conditions. Carbonate production again dominates, and the slope progrades and aggrades over its failed deposits. Changes in dip attitude across SDS suggest a complex lobate morphology as the slope builds over its relict topography (Figs. 1016). This style of progradation and aggradation of carbonate slopes has been described elsewhere in the Bone Spring Formation (Saller et al., 1989) and in other carbonate clinoform systems (Sonnenfeld, 1991; Gómez-Pérez et al., 1999; Katz et al., 2010; Playton et al., 2010; Playton and Kerans, 2018). Deformed (FA2) and channelized (FA3) facies are common at the base of clinothems, as the relict scarp surfaces attract coarse-grained sediment gravity flows (Eggenhuisen et al., 2010; Janson et al., 2011; Stevenson et al., 2015). Toward the top of clinothems, undeformed lime mudstones (Lithofacies 1) dominate as the slope finds local equilibrium (e.g., shallowing dips in Clinothem 8, see Fig. 14). SDS 4 represents a surface associated with this A/S shift.

Seven of the nine detachment surfaces (SDS 1, 2, and 5–9) in the study area likely followed time steps ABD, with no major terrigenous influx. From SDS 3–4, the system likely followed an ABCD path, with a large decrease in A/S accounting for a larger flux of terrigenous sediment. A prolonged decrease in A/S (e.g., the Bone Spring 1st, 2nd, and 3rd Sands) would follow a similar ABCD path, with time step C representing relatively long periods with large volumes of terrigenous sediment bypass to the basin (cf. Stevenson et al., 2015). A schematic cross section of a hypothetical time sequence (i.e., ABDABCD) is illustrated in Figure 18E.

6.2 Implications for Sequence Stratigraphy

Sequence stratigraphic concepts are commonly used to predict facies from seismic-scale geometries (Mitchum et al., 1977; Vail, 1987). However, allogenic forcing is often overly relied upon without considering the effects of autogenic forcing and along-strike variability (see discussion in Burgess, 2016), resulting in over-simplified stratigraphic “pancake” models for basin fill (e.g., a local sand body interpreted to represent a correlatable lowstand sand “sheet” across a basin; Saller et al., 1989; Montgomery, 1997b; Nance and Rowe, 2015; Crosby et al., 2018; Bhatnagar et al., 2018; Schwartz et al., 2018). In reality, as many studies have shown, sediment supply, accommodation, along-strike variability, and other factors may affect the regional and local development of both low-order and higher-order systems tracts and sequences (Covault et al., 2007; Burgess, 2016; Madof et al., 2016; Harris et al., 2018; Trower et al., 2018; Chiarella et al., 2019).

Results from this study provide insight into different forcing mechanisms that result in carbonate and terrigenous sediment mixing and partitioning. From an allogenic perspective, terrigenous sediment associated with slope detachment surfaces (Figs. 3B and 17) may record relative sea-level fluctuations of variable magnitude. In such cases, similar processes would be expected to occur regionally, resulting in a relatively correlatable basin stratigraphy (Li et al., 2015; Nance and Rowe, 2015). In the study area, the terrigenous sediment of Clinothem 4 is discontinuous and interbedded with carbonate sediment (Fig. 3B), suggesting widespread correlability is unlikely. Furthermore, poor age control prevents the slope-to-basin correlation of SDSs and clinothems, and thus it is difficult to prove or deny relative sea level as a causal mechanism for the observed stratigraphic architecture in the study area. From an autogenic perspective, variations in rates of progradation and aggradation of carbonates result in a rugose margin, as recognized in the subsurface Permian Basin (Saller et al., 1989) and in modern-day carbonate margins (Mulder et al., 2012). This rugosity provides conduits for transport of both coarse-grained carbonate and terrigenous sediment to the basin without invoking relative sea-level change (cf. Boyd et al., 2008). As the margin builds by episodes of growth and failure (Fig. 18; Saller et al., 1989; Playton et al., 2010), along-strike topographic variability may result in local differences in sediment input, clinothem composition and architecture (cf. Madof et al., 2016), forming a heterogeneous basin stratigraphy with contemporaneous carbonate and terrigenous deposition (Fig. 18E). The rugosity of the Bone Spring slope in Shumard Canyon (Fig. 10B), and the presence of (1) terrigenous sediment (Clinothem 4, Fig. 12) and (2) stacked submarine-channel deposits in Clinothems 7–8 (Fig. 10A) suggests that the Shumard Canyon area may have been an entry point for coarse-grained sediment carried to a portion of the northwestern Delaware Basin (Fig. 1A inset). Other sediment conduits identified in both the Cutoff (Hurd et al., 2018) and Brushy Canyon (Gardner et al., 2008) formations at this location corroborate a persistent basin entry point in this area.

6.2.1 1st Bone Spring Sand Exposed in Shumard Canyon

In the study area, the Bone Spring slope is >90% carbonate-dominated, but there is a significant accumulation of terrigenous sediment in Clinothem 4 (Figs. 12, 16A, and 17A). This package is thickest and best developed in Shumard Canyon and becomes thin and discontinuous downslope (Figs. 3B, 10, 16A, and A2). The volume and proportion of terrigenous sediment (predominantly quartz silt) in this location and its stratigraphic position, suggest that SDS 3 is the L5 sequence boundary and that Clinothem 4 is therefore the basal unit of the L5 sequence of Fitchen (1997), commonly referred to as the “1st Bone Spring Sand” in the basin (Fig. 2). Alternatively, Clinothem 4 may represent a localized area of terrigenous sediment delivery within the L5 sequence, but not the basal unit. In either case, the lithologic and architectural heterogeneity of Clinothem 4 suggests that autogenic and allogenic processes acted concurrently to build Bone Spring stratigraphy. Deconvolving those superimposed signals would be difficult using lithostratigraphy alone, so further work revising biostratigraphy and outcrop-to-well-log correlations is required to support our interpretations. If Clinothem 4 is indeed equivalent to the1st Bone Spring Sand, the observed lateral lithological heterogeneity will be important when performing local and regional well-to-well correlations in the Delaware Basin and in similar mixed sediment routing systems (Hampson, 2016; Madof et al., 2016; Romans et al., 2016).

6.3 Sub-Seismic Scale Predictions from Seismic-Scale Architectural Elements

Slope detachment surfaces 1–9 can be correlated for more than 1 km and have relief/thickness values greater than 20 m, indicating that they have geometries comparable to seismic data in the Permian basin (Fig. 19). The spatial and temporal distribution of facies and depositional elements documented by this study demonstrates how sub-seismic facies variability can be tied to seismic-scale architecture. Subsurface features of similar scale and architecture to SDSs are imaged in seismic-reflection data from the Leonardian margin along the Northwest Shelf (Fig. 19A; Sarg, 1988; Sarg et al., 1999). A seismic-scale basinal siliciclastic wedge is interpreted (labeled Lower Avalon, Fig. 19B) with a carbonate package prograding over the top of the sand (labeled Victorio Peak and Bone Spring Carbonate, Fig. 19B). Clinoform geometries are identified within the prograding package (orange lines, Fig. 19B). Outcrops of the Bone Spring Formation are shown at the same scale as the seismic data (Figs. 19C1 and 19C2). The SDS may represent larger clinoform geometries, whereas the terrigenous-rich clinothem may be the slope expression of a terrigenous fan complex (e.g., 1st Bone Spring Sand) in the basin. We expect that within a clinothem, MTDs and terrigenous lithologies will occur near the base of the clinothem and will onlap slope detachment surfaces near the toe-of-slope, but will become progressively more carbonate-rich toward the top (Fig. 18E), potentially aiding facies interpretations from seismic data.

7. CONCLUSIONS

Outcrops of the Bone Spring Formation of Guadalupe Mountains National Park provide an opportunity to investigate slope-building processes and sediment distribution in a mixed carbonate-siliciclastic margin. Slope-building clinothems of mixed lithology are bounded by slope detachment surfaces reflecting large-scale subaqueous failure of the carbonate margin. Terrigenous sediment associated with slope detachment surfaces suggests that slope failure may be positively linked to terrigenous sediment influx. At the base of typical clinothems, carbonate mass-transport deposits and submarine-channel fills are common as the slope fills the topography created by failure. By contrast, in the upper portions of clinothems, undeformed carbonate mudstones were deposited as the slope found local equilibrium. Bedding attitudes are significantly different across slope detachment surfaces, suggesting that the primary mechanism in slope evolution were repeated mass wasting and infill resulting in a rugose margin. An observed slope inflection point contains abundant evidence of failures and of submarine channel deposits, suggesting that coarse-grained entry points to the basin were influenced by slope morphology.

This study also provides insight into sequence stratigraphic concepts in a mixed-lithology system. A terrigenous-rich clinothem in Shumard Canyon is interpreted as the slope equivalent of the basinal 1st Bone Spring Sand. However, variability in the lithologic stacking patterns and lateral continuity of this clinothem suggests that both autogenic and allogenic processes influenced deposition. This complexity has important implications for well-to-well correlations in the Delaware Basin. The slope-building processes documented in the Bone Spring Formation acted as a primary control on sediment distribution, stacking patterns, and depositional styles, which can be utilized to predict reservoir-forming facies in the basinal Delaware Basin. Insights from this study can also be utilized to aid in reconstructing the evolution of other mixed-lithology margins globally.

ACKNOWLEDGMENTS

We would like to thank Colorado School of Mines and the Chevron Center for Research Excellence (CoRE) for providing the primary funding for this project. Additional funding was provided by the West Texas Geological Society (WTGS), American Association of Petroleum Geologists Grants-In-Aid, and the Society for Sedimentary Geology (SEPM) Foundation. A special thank you to Jonena Hearst and Guadalupe Mountain National Park for access to the Western Escarpment. We would also like to thank Sebastian Cardona, Evan Gross, Thomas Martin, and Enry Horas Sihombing for assistance during field work and Mary Carr, Jenn Pickering, Domenico Chiarella, Bill Fitchen, Xavier Janson, Greg Hurd, and two anonymous reviewers for providing useful feedback that improved the paper.

1Supplemental Material. Figure S1: Location of six photopanels compiled along the outcrop overlaying the 3D textured model. Figure S2: West-facing outcrop between Shumard Canyon and Bone Canyon. Characteristics of slope detachment surfaces aid in correlating surfaces from Shumard to Bone Canyons. Note the discontinuous nature of the terrigenous sediment within Clinothem 4 from Shumard to Bone Canyons (also shown in Fig. 3B). Figure S3: Location of detailed outcrop photos shown in this publication. Please visit https://doi.org/10.1130/GEOS.S.14344085 to access the supplemental material, and contact editing@geosociety.org with any questions.
Science Editor: David E. Fastovsky
Associate Editor: Gregory D. Hoke
Gold Open Access: This paper is published under the terms of the CC-BY-NC license.