KGS Home Geology Home Start of Sedimentary Modeling book

Kansas Geological Survey, Subsurface Geology 12, p. 67-70

Depositional-sequence analysis and computer simulation of Upper Pennsylvanian (Missourian) strata in the midcontinent United States

W. Lynn Watney, John A. French, and Jan-Chung Wong
Kansas Geological Survey

Upper Pennsylvanian (Missourian) mixed carbonate-siliciclastic depositional sequences in the upper midcontinent United States accumulated on ramp and platform settings along margins of the episodically subsiding Arkoma and Anadarko basins. These distinctive, often cyclical deposits are similar to those that accumulated in many parts of the world during the late Paleozoic (Ross and Ross, 1988). Varied allogenic and autogenic causal mechanisms have been proposed to explain these successions; these include glacial eustasy (Heckel, 1986) and tectonic subsidence (Klein and Willard, 1989). Regional subsurface studies in Kansas indicate that changing shelf configuration (differential relief and tectonic subsidence) and eustasy were both important in generation of the Missourian sequences (Watney, 1985; Watney et al., 1988). Computer modeling coupled with sequence- stratigraphic analysis will more rigorously address the relative importance of the causal mechanisms.

The Missourian depositional sequences include several genetic units: a basal flooding unit, a condensed section, a shallowing-upward unit, and, on the middle and upper shelf, regionally significant subaerial exposure surfaces (Watney et al., 1988). The components of these Missourian depositional sequences on the upper carbonate platform closely parallel the lithofacies of the transgressive-regressive "Kansas cyclothem" defined by Heckel (1986), and fifth-order transgressive-regressive cycles of Busch and Rollins (1984). The genetic units within a depositional sequence can vary considerably between siliciclastic-dominated and carbonate-dominated depositional systems in shelf and basin settings. Genetic units can be extended to the concept of systems tracts (Vail, 1987) as information is obtained on stratal geometries and facies through subsurface and surface investigations.

The computer model is a forward (process-response) stratigraphic simulation with input parameters that include sea-level elevation, tectonic subsidence, sedimentation rate, and elevation of the depositional surface. The modular Turbo Pascal program contains special algorithms that modify rate of sedimentation as a function of rate of change of sea level or water depth.

A glacial-eustatic sea-level curve has been approximated for the Missourian using a modified version of the del-O18 curve of the Late Pleistocene. The average period of major cycles used is 400 ka, the longest period suggested by Heckel (1986). A maximum amplitude of 110 m (363 ft) is used, which is similar to that of the Pleistocene. Sediment-surface elevations used in modeling range over 50 m (165 ft), as estimated from regional subsurface studies of the northern midcontinent depositional sequences in western Kansas (Watney, 1985).

Output consists of a plot of time versus depth showing the elevations of sea level and the sediment surface, and the subsidence path (fig. 1A). Water depth, sediment thickness, and distinctive genetic units (including flooding units, condensed sections, shallowing-upward units, and subaerial-exposure horizons) can be examined and compared with observed strata. The output also portrays sequence development versus time which is useful in restoring the sequences at various time steps (fig. 1B) or in construction of Wheeler diagrams. Model results are sought which closely reflect empirical, and at this stage of our understanding, conceptual models of shelf sequences (fig. 1C).

Figure 1--A) Example of a forward computer model of mid-shelf Pennsylvanian carbonate-dominated depositional sequences deposited in a lower shelf setting. Simulated stratigraphic sequences consist of water deposits, <3 m (10 ft; dot pattern), below wavebase (brick), flooding units (T's), and condensed section (C). Sea-level curve is derived from expansion of Pleistocene del-O18 cycles to 400,000 years. Presentation is both in depth (reference elevations in meters) along the vertical axis, and time in thousands of yeras along the horizontal axis; B) Depth cross section defined by elevations derived for simulated sequence #1 including information from model shown in (A). Section shows top lap between upper and lower shelf position and more "high fidelity" record in lower shelf vertical sequence; C) Conceptual model for Dennis sequence for western Kansas showing extent of sequence from landward limit to sediment-starved basinward limit in western Anadarko basin.

Example of a forward computer model.

Top figure is cross section from computer model, bottom figure is model of Dennis sequence.

Features of the simulated carbonate-dominated depositional sequences that resemble the rock record include 1) thicker sequences toward the basin margins that are characterized by increasingly higher preservation due to lower elevation and longer periods of sediment accumulation, which results in increased resolution of the sedimentary record; 2) low average sediment accumulation rates of the sequences compared to modern sedimentation rates, largely because sediment accumulation is interrupted by significant hiatuses associated with condensed sections and subaerial-exposure surfaces; compaction and dissolution no doubt play a role as well; 3) complex successions of units occur within a depositional sequence due to rapid temporal changes in sediment accommodation, resulting, for example, in repeated genetic units within one depositional sequence; and 4) progressive loss in elevation of the shelf through high subsidence rates coupled with rapid fluctuating sea level results in sustained sediment starvation for parts of the section.

Inverse modeling (transformation of stratigraphic column to processes versus rate profiles) will be required to interpret and model specific sequences observed in the rock record. Coherent signals in estimated rates and magnitude of sedimentary parameters versus time obtained from inverse modeling can be potentially extracted through mathematical and statistical analysis.

Preliminary comparison of two study areas, one in western Kansas and the other in eastern Kansas, reveals contrasting styles of sedimentation and challenges to sequence analysis and modeling. Western Kansas was the site of a broad, carbonate-dominated ramp adjacent to the Anadarko basin, then an active foreland basin. A succession of thick (up to 25 m [83 ft]) oolitic grainstones accumulated on a ramp created by long-term, but episodic, differential tectonic subsidence during the early Missourian (Watney, 1985). The ramp setting reverted occasionally to a platform due to episodic waning subsidence. Progradation of the shoal-water carbonate facies over the ramp formed stacked sedimentary wedges that are exemplified by the Dennis and Swope sequences in the lower Kansas City Group, with each extending over 150 km (90 mi) across the shelf. Estimated relief across this shelfward portion of the ramp amounted to at least 15 m (50 ft; 0.1 m/km [0.3 ft/mi]). A minimum progradation rate of 0.38 m (1.25 ft)/ka is suggested for a 400,000-year sequence period.

A considerable portion of the relief on the ramp was apparently filled in by the wedges of shoal-water carbonates. However, subaerial exposure and an ensuing depositional hiatus that terminated the sequence apparently led to reestablishment of ramp conditions due to increased subsidence toward the Anadarko basin in southwestern Kansas. Estimated average subsidence rates during the Missourian range from less than 0.05 m (0.17 ft)/ka in southern Nebraska to greater than 0.20 to 0.30 m (0.7-1 ft)/ka along the margin of the Anadarko basin in southern Kansas (Kluth, 1986). Computer simulations support the feasibility of these subsidence rates.

The geometries of depositional sequences in the lower Kansas City Group in eastern Kansas appear to be strongly influenced by preexisting depositional topography along margins of broad platform and localized delta lobes (Heckel et al., 1985; Heckel, 1988; French, 1988). Relief and gradient of slopes associated with this topography range from 25 to 30 m (83-99 ft) over a distance of 10-30 km (6-18 mi; 2.5 to 3 m/km [8.3-10 ft/mi] or 25 to 30 times the slope of the western Kansas shelf). While oolitic grainstones are the dominant facies in the progradational wedge in western Kansas, phylloid algal carbonate buildups dominate the narrower progradational or aggradational wedges in many of the cycles in eastern Kansas. In southern Kansas, along the margin of the Arkoma basin, alternating siliciclastic and carbonate aggradation/progradation episodically extended the shelf margin southward beyond an original shelf margin position that occurred along a relatively steep (as compared to western Kansas), basinward-directed slope on the edge of a delta platform in the Pleasanton Group. Subsidence during the early Kansas City interval was minimal during development of the Sniabar and Swope depositional sequences. Increased subsidence during the succeeding Dennis Sequence led to backstepping of the shelf margin over underlying shelf deposits, facilitating accumulation and preservation of apparent parasequences. Computer simulations recreate basinal and lower shelf conditions that correspond to these scenarios by utilizing the best available estimates of sedimentary parameters.

Problems that remain to be resolved in Missourian sequences include establishing the details of internal stratigraphy and obtaining better estimates of accumulation rates, durations, and water depths of sediment accumulation.

Refined computer simulation, coupled with sequence-stratigraphic analysis, will become a tool that can substantially improve our understanding of the controlling parameters on deposition and straw geometries. Modeling will be useful in constraining and testing these parameters and in increasing the accuracy and precision of geologic interpretations. Once more precise modeling parameters are known, simulations will become useful in the prediction of sequence characteristics related to mineral-resource appraisal.


Busch, R. M., and Rollins, H. B., 1984, Correlation of Carboniferous strata using a hierarchy of transgressive-regressive units: Geology, v. 12, p. 471-474

French, J. A., 1988, Genetic sequence analysis of midcontinent cyclothems--detemiination and importance of autogenic and allogenic depositional controls: Geological Society of America, Abstracts with Programs, v. 20, no. 2, p. 99

Heckel, P. H., 1986, Sea-level curve for Pennsylvanian eustatic marine transgressive-regressive depositional cycles along midcontinent outcrop belt, North America: Geology, v. 14, p. 330-334

Heckel, P. H., 1988, Relations of basinward stratigraphy of midcontinent Pennsylvanian cyclothems to amount of sea-level drop during regression: Geological Society of America, Abstracts with Programs, v. 20, no. 2, p. 101-102

Heckel, P. H., Harris, J., and Watney, W. L., 1985, Recent advances in interpretation of Late Pennsylvanian cyclothems, Guidebook for Midcontinent SEPM Field Trip, October 12; in, Proceeding of Third Annual Meeting and Field Conference: Society of Economic Paleontologists and Mineralogists, Midcontinent Section, p. 23-69

Klein, George de V., and Willard, D. A., 1989, Origin of Pennsylvanian coal-bearing cyclothems of North America: Geology, v. 17, p. 152-155

Kluth, C. F., 1986, Plate tectonics of the ancestral Rocky Mountains; in, Paleotectonics and Sedimentation in the Rocky Mountain Region, United States, J. A. Peterson (ed.): American Association of Petroleum Geologists, Memoir 41, p. 353-369

Ross, C. A., and Ross, J. R. P., 1988, Late Paleozoic transgressive-regressive deposition; in, Sea-Level Changes--An Integrated Approach, C. K. Wilgus, B. S. Hastings, C. G. St. C. Kendall, H. M. Posamentier, C. A. Ross, and J. C. Van Wagoner (eds.): Society of Economic Paleontologists and Mineralogists, Special Publication 42, p. 227-247

Vail, P. R., 1987, Part 1--Seismic stratigraphy interpretation procedure; in, Atlas of Seismic Stratigraphy, A. W. Bally (ed.): American Association of Petroleum Geologists, Studies in Geology 27, v. 1, p. 1-10

Watney, W. L., 1985, Resolving controls on epeiric sedimentation using trend surface analysis: Mathematical Geology, v. 17, p. 427-454

Watney, W. L., Knapp, R. W., French, J. A., Jr., and Doveton, J. H., 1988, Application of sequence-stratigraphic analysis to thin cratonic carbonate-dominated shelf cycles (Upper Pennsylvanian) in the midcontinent (abs.): American Association of Petroleum Geologists, Bulletin, v.72, p. 257

Kansas Geological Survey
Comments to
Web version May 12, 2010. Original publication date 1989.