Quantitative Constraints on the Origin of Stratigraphic Architecture at Passive Continental Margins: Oligocene Sedimentation in New Jersey, U.s.a

The Oligocene of the New Jersey continental margin is divisible into as many as eight sequences and 23 lithofacies associations, documented in a series of seven boreholes across the modern coastal plain. This paper summarizes the sequence architecture of these deposits, interpreted from high-resolution biostratigraphy and Sr-isotope chemostratigraphy, and evaluates the factors that governed patterns of sedimentation, making use of previously published quantitative estimates of water-depth changes and eustasy from 2-D foraminiferal paleoslope modeling and flexural backstripping. Each sequence is markedly wedge-shaped, thinning both landward of the rollover in the underlying sequence boundary (the point at which the surface steepens into a clinoform), and seaward of the rollover in the overlying boundary. Each bounding surface is associated with evidence for offlap–onlap geometry and at least locally with benthic foraminiferal evidence for abrupt upward shoaling. Most unconformities merge up dip into a single surface marking the Oligocene–Miocene boundary. Earliest Oligocene unconformities (33.5–31.6 Ma) merge downdip as a result of sediment starvation on the deep shelf. Conventional lithostratigraphic units within the New Jersey Oligocene are highly diachronous. For example, the base of Atlantic City Formation at Cape May (a downdip borehole) is at least 6.6 Myr younger than the top of the same formation at ACGS#4 (an updip borehole). Factors controlling patterns of sedimentation include: (1) a terraced physiography, with gradients ranging from 1:1,000 (0.068) on the coastal plain and shallow shelf and 1:500 (0.118) on the deep shelf to , 1: 100 (1.08) on an intermediate slope; (2) generally low siliciclastic sediment flux, with in situ production of authigenic glauconite, especially during times of transgression; (3) a location landward of the hinge zone of the passive margin, with slow tectonic subsidence augmented by compaction and sediment loading; (4) low to moderate amplitudes and rates of eustatic change (10–50 m over spans of ; 1–2 Myr); and (5) an active wave climate that permitted efficient lateral transport and complete bypass of sediment at paleodepths of at least 20 6 10 m. Sequence architecture in the New Jersey Oligocene differs from that of the standard ‘‘Exxon model.’’ Sequences are highstand-dominated, in spite of deposition and preservation largely seaward of the rollover in each underlying sequence boundary. Transgressive systems tracts are thin. Recognizable lowstand units did not form because efficient transfer of sediment across the shallow shelf, combined with the absence of major river systems in the area of study, prevented the reorganization of sedimentation patterns commonly associated with point-source development, in spite of rates of eustatic fall considerably greater than the local rate of tectonic subsidence. Repeated eustatic rises and falls are expressed primarily by variations in paleo-water depth. Although ; 65–80% of the shallow shelf that had been flooded during each rise became subaerially exposed during the subsequent fall, well developed offlap at each sequence boundary is due primarily to marine bypassing and degradation rather than to ‘‘forced regression.’’ Sequence boundaries correspond in time at their correlative conformities not with the onset of falling ‘‘relative’’ sea level, but with the start of eustatic rise. INTRODUCTION The roles of various factors in governing sedimentation patterns at continental margins have been debated since the early days of modern geology (e.g., Suess 1906; Stille 1924). Of these factors, tectonics, eustasy, and sediment supply have stood out as amongst the most important—tectonics for ultimately making the space needed for sediment to accumulate, and all three factors for potentially influencing the manner in which available space is filled (e.g., Burton et al. 1987; Posamentier et al. 1988; Galloway 1989; Reynolds et al. 1991; Underhill 1991; Plint et al. 1993; ChristieBlick and Driscoll 1995). The emergence of seismic and sequence stratigraphy in the 1970s and 1980s led to the development of new concepts about facies arrangements and stratigraphic architecture and their possible relation to eustatic change (e.g., Vail et al. 1977; Vail et al. 1984; Vail et al. 1991; Haq et al. 1987; Vail 1987; Plint 1988, 1993; Posamentier et al. 1988; Posamentier et al. 1992; Sarg 1988; Van Wagoner et al. 1990; Carter et al. 1991; Christie-Blick 1991; Hunt and Tucker 1992; Karner et al. 1993; Posamentier and James 1993; Schlager 1993; Helland-Hansen and Gjelberg 1994; Christie-Blick and Driscoll 1995; Van Wagoner 1995; Naish and Kamp 1997; Posamentier and Allen 1999; Plint and Nummedal 2000; Posamentier and Morris 2000). The eustatic paradigm has been highly influential, but even as evidence for strong eustatic forcing during times of continental glaciation has solidified (e.g., Naish and Kamp 1997; Miller et al. 1998a), questions have persisted about precisely how patterns of sedimentation respond to changing sea level. In the absence of quantitative stratigraphic constraints, interpretations have largely been qualitative and inseparable from the loosely specified concept of relative sea-level change (Posamentier et al. 1988; Posamentier and Allen 1999). Stratigraphic studies of core samples from Oligocene sediments of the New Jersey coastal plain have yielded a remarkably well calibrated record of changing facies and paleo-water depths in eight unconformity-related sequences, from which it has been possible to extract a unique quantitative interpretation of eustatic change on a million-year timescale (Pekar et al. 2000; Pekar et al. 2001; Kominz and Pekar 2001; Pekar and Kominz 2001). Our previously published articles on these sediments focus on high-resolution chronostratigraphy using biostratigraphy (planktonic foraminifers, nannofossils, diatoms, and dinocysts) and Sr-isotope chemostratigraphy; on the development of a quantitative methodology for estimating water-depth changes in two dimensions; and on using flexural backstripping to place constraints on eustatic change. This paper takes stock of how the sequences are put together, and reexamines the factors responsible. SEQUENCE STRATIGRAPHIC INTERPRETATION The Oligocene sequence stratigraphy of New Jersey was interpreted from a series of boreholes projected onto a transect across the modern coastal plain (Figs. 1–3; Pekar 1999; Pekar et al. 2000). Sequence boundaries and systems tracts were delineated on the basis of inferred stratal geometry and facies arrangements, and without reference to sea-level change (see Christie-Blick 1991, 2001; Christie-Blick and Driscoll 1995). This distinction, which is consistent with the way in which systems tracts were first defined 228 S.F. PEKAR ET AL. FIG. 1.—Location map. The southern part of New Jersey, eastern United States, is shown with locations of sites used in this study: Island Beach, Atlantic City, Cape May (Leg 150X boreholes); Bass River (Leg 174AX borehole); AMCOR 6011, ACGS#4, Great Bay and Jobs Point (U.S.G.S. onshore and offshore wells). Dip section A–A9 is drawn perpendicular to Cretaceous outcrops; strike lines are projected from boreholes onto that section. Updip part of Oceanus 270 line 529 is shown with Cape May and Atlantic City sites projected along strike onto that line. Circled numbers 1 and 2 represent location of rollovers for seismic surface m6 (Oligocene–Miocene boundary) and the immediately underlying sequence boundary (Monteverde et al. 2000). FIG. 2.—Distribution of New Jersey Oligocene sequences projected onto dip section A–A9 (see Fig. 1) at ; 24 Ma. Sequences ML and O1 through O6 are of Oligocene age. Ties for reconstructed sequence boundaries are depths at each borehole. Clinoforms are required by the data; the sigmoidal shapes are conjectural. Clinoform relief is inferred from twodimensional flexural backstripping (Kominz and Pekar 2001). Bold line indicates original depth and gradient (1/500) of Eocene–Oligocene surface. Bold dashed line indicates paleoshelf gradient landward of rollover (1/1000). (Brown and Fisher 1977; Vail 1987), is necessary to avoid circularity in determining how stratigraphy relates quantitatively to eustasy. In spite of some terminology currently in use (highstand, lowstand, falling stage, forced regressive, etc.; e.g., Vail 1987; Hunt and Tucker 1992; Posamentier et al. 1992; Helland-Hansen and Gjelberg 1994; Naish and Kamp 1997; Plint and Nummedal 2000; Posamentier and Morris 2000), and the apparent intent of some authors, in this paper systems tracts are specifically not interpreted according to whether sea level is thought to have been high, low, or falling, etc. (see below for further discussion). While better terms might be considered, and uncertainties exist in practice about the precise location, continuity, and time significance of systems-tract boundaries, we find the threefold subdivision of sequences into lowstand, transgressive, and highstand systems tracts (Vail 1987) more useful and potentially less subjective than the several variants that have emerged in the past decade. Vail’s scheme requires only an assessment of whether the shoreline was moving generally seaward (highstand and lowstand) or landward (transgressive), and of stratigraphic location with respect to the transition from offlap beneath a sequence boundary (highstand) to onlap above (lowstand and transgressive). The transition from highstand to lowstand sedimentation requires a change in the pattern of progradation and, in terrigenous systems, generally begins with the development of point sources. Lowstand systems tracts defined in this way are not present in most sequences, even those overlying prominent unconformities. We note the ill-advised usage by others of the term lowstand for coarse-grained and/or nonmarine lithosomes overlying sequence boundaries, whether or not continued seaward movement of the shoreline can be demonstrated (e.g., Van Wagoner 1995); and for offlapping stratigraphic elements below geometrically delineated sequence boundaries (e.g., Hunt and Tucker 1992; Helland-Hansen and Gjelberg 1994; Posamentier and Allen 1999; Posamentier and Morris 2000). The transition from onlap to offlap is typically associated with the highstand systems tract (Christie-Blick 1991), although the degree to which offlap is developed is highly variable. In some cases, a combination of limited sediment accumulation and degradation beneath a developing surface results in the erosional truncation of already deposited highstand and transgressive units or even the amalgamation of two or more unconformities (e.g., Kidwell 1997). In our interpretation of the New Jersey Oligocene, each boundary of each sequence is represented by an unconformity that either passes basinward into a correlative conformity, where the associated hiatus is no longer resolvable, or amalgamates with another sequence boundary in an interval of sediment starvation (condensed section). Sequence boundaries are associated with evidence for offlap–onlap geometry (from a comparison of high-resolution chronology in adjacent boreholes), and at least locally with benthic foraminiferal evidence for abrupt upward shoaling, a characteristic feature of this kind of surface (Christie-Blick 1991, 2001). In continuously cored boreholes, unconformities are typically associated with an irregular erosional surface, a marked change in lithofacies and benthic foraminiferal biofacies, hardground development, and a sharp upward increase in gamma-ray log response (features related at least in part to initial marine flooding). In boreholes that were cored discontinuously (at intervals of 5 or 10 feet), unconformities are commonly not recovered but are instead bracketed by samples suggesting the presence of a hiatus and revealing contrasts in lithofacies and/or biofacies. In these cases, the locations of surfaces are interpreted from gamma-ray log data. Condensed sections are typically 229 CONSTRAINTS ON STRATIGRAPHIC ARCHITECTURE AT PASSIVE CONTINENTAL MARGINS FIG. 3.—Distribution of New Jersey Oligocene sequences and borehole locations projected onto dip section A–A9 (see Fig. 1), with datum at base of sequence Kw1a. Depths are in meters. Ages of sequences: E10 and E11 are latest Eocene (Browning et al. 1997); ML and O1 to O6 are Oligocene (Pekar et al. 2000); and Kw0 and Kw1a are earliest Miocene (Miller et al. 1997). Successive sequences are arranged laterally, with the oldest landward and the youngest seaward. Also shown are lithology, ages of strata immediately below and above sequence boundaries, and lithostratigraphy (from Pekar et al. 1997b). Sequences O2 and O5 are shaded to emphasize correlations between boreholes. Lithologic key applies also to Figures 4–7 and 11–13. marked by at least one of the following: high concentrations of authigenic glauconite sand (an indicator of low terrigenous input; McRae 1972); abundant benthic foraminifers with peak species abundances of uvigerinids; and a change from upward-deepening to upward-shallowing trends (Pekar 1999; Pekar and Kominz 2001). They are intervals of sediment starvation. With few exceptions, maximum flooding surfaces cannot be recognized objectively in our data. Lithofacies and Age Control The strata have been divided into 23 lithofacies associations on the basis of grain size, mineral abundance (mainly quartz and glauconite), whether the glauconite is in situ or detrital (reworked or transported), diagnostic microfauna, and the presence of shells and associated sedimentary structures (Table 1; summarized from lithologic descriptions in Miller et al. 1994; Pekar et al. 1997a; Pekar 1999). In situ glauconite typically forms in quiescent, sediment-starved, low-oxygen middle neritic and deeper paleoenvironments (McRae 1972). Detrital glauconite is suggested by: (1) abraded, cracked, and broken grains; (2) mixed populations of green and brown grains (weathered to goethite); (3) an association with abundant quartz; (4) an association with inner neritic benthic foraminiferal taxa (Pekar et al. 1997a). Cumulative weight percentage data were collected for the medium to coarse and fine quartz sand fractions, silt–clay, glauconite sand, and shell material. (For the purpose of simplification in this paper, the term fine sand includes fineand very fine-grained sand on the Wentworth scale, or 63 to 250 mm; and the term coarse sand includes coarseand very coarse-grained sand on the Wentworth scale, or 500 to 2,000 mm.) In intervals with glauconite, abundances of other components were visually estimated from the greater than 63 mm size fraction. In intervals without glauconite and shell material, the percentages of fine versus medium to coarse sand were obtained by dry sieving and weighing. An age model was developed for New Jersey Oligocene strata by integrating planktonic foraminiferal, dinocyst, diatom, and nannofossil biostratigraphy, Sr-isotopic chemostratigraphy, and limited magnetostratigraphy 230 S.F. PEKAR ET AL. TABLE 1.—Summary of late Paleogene (34.2–23.9 Ma) lithofacies in New Jersey. Lithofacies Code Description Type Location Depositional Environment Medium to coarse quartz association C1 Dark olive gray (5GY 4/1) coarse to gravelly glauconitic (10–20%) quartz sand; massive, microfossils are sparse. Uppermost sequence O6 at Cape May Inner neritic C2 Dark greenish gray (5GY 4/1) to olive gray (5Y 4/2) slightly glauconitic (,10%), shelly, medium to coarse quartz sand. Microfossils are sparse to absent; typically massive. Sequence O5 at Atlantic City Inner to inner middle neritic C3 Dark greenish gray (5GY 4/1) to dark green (5GY 4/1), glauconitic (10–30%), medium to coarse quartz sand, mostly massive with occasional thin parallel bedding. Microfossil are sparse to moderate. Sequence O6 at Cape May Inner neritic C4 Light gray (5Y 7/1) to gray (5Y 6/1) medium to coarse quartz sand, shells present, microfossils are sparse to absent. Sequence O1 at ACGS#4; Sequence O4 at AMCOR Inner neritic C5 Olive (5Y 4/3) slightly glauconitic (,10%) medium to coarse quartz, with abundant shell fragments, abundant foraminifers. Sequence O5 at Great Bay Inner to middle neritic